Building Turbopuffer: Gergely Orosz (@pragmaticengineer ) × Simon Eskildsen (CEO)

All right. All right. It's great to be here and today with me It's great to be here and today with me It's great to be here and today with me here I'm I'm Gerge, author of the here I'm I'm Gerge, author of the here I'm I'm Gerge, author of the pragmatic engineer and I'm excited to pragmatic engineer and I'm excited to pragmatic engineer and I'm excited to have a chat with Simon Ericson uh have a chat with Simon Ericson uh have a chat with Simon Ericson uh founder CEO of Turbopuffer a very founder CEO of Turbopuffer a very founder CEO of Turbopuffer a very technical CEO and we're going to have a technical CEO and we're going to have a technical CEO and we're going to have a pretty technical discussion but before pretty technical discussion but before pretty technical discussion but before we jump into it Simon I wanted to ask we jump into it Simon I wanted to ask we jump into it Simon I wanted to ask where did you fall in love with where did you fall in love with where did you fall in love with computers? computers? computers? um through PowerPoint. um through PowerPoint. um through PowerPoint. >> PowerPoint >> PowerPoint >> PowerPoint you I don't know if any of you know this you I don't know if any of you know this you I don't know if any of you know this but in power well you probably know this but in power well you probably know this but in power well you probably know this but in PowerPoint right you can make the but in PowerPoint right you can make the but in PowerPoint right you can make the the diagrams and stuff when you click the diagrams and stuff when you click the diagrams and stuff when you click them go to another slide them go to another slide them go to another slide that becomes touring complete real quick that becomes touring complete real quick that becomes touring complete real quick right you can sort of you know create right you can sort of you know create right you can sort of you know create very complicated convoluted games and very complicated convoluted games and very complicated convoluted games and then at some point you know you make it then at some point you know you make it then at some point you know you make it through the Microsoft Office suite and through the Microsoft Office suite and through the Microsoft Office suite and you discover Front Page. Do you remember you discover Front Page. Do you remember you discover Front Page. Do you remember Front Page? Front Page? Front Page? >> Yeah, I remember Front Page. It it it >> Yeah, I remember Front Page. It it it >> Yeah, I remember Front Page. It it it was supposed to eliminate the need for was supposed to eliminate the need for was supposed to eliminate the need for all any front-end developers. all any front-end developers. all any front-end developers. >> Exactly. And it it only worked in >> Exactly. And it it only worked in >> Exactly. And it it only worked in Internet Explorer. I remember a Internet Explorer. I remember a Internet Explorer. I remember a heartbreak I had one day when someone heartbreak I had one day when someone heartbreak I had one day when someone opened a website I created in Firefox opened a website I created in Firefox opened a website I created in Firefox and it just it was it was all over the and it just it was it was all over the and it just it was it was all over the place. And then one day I accidentally place. And then one day I accidentally place. And then one day I accidentally clicked the HTML clicked the HTML clicked the HTML thing in front page and it just showed thing in front page and it just showed thing in front page and it just showed all of this stuff that I couldn't make all of this stuff that I couldn't make all of this stuff that I couldn't make sense of and it just started looking at sense of and it just started looking at sense of and it just started looking at it and then going online and finding it and then going online and finding it and then going online and finding little snippets that you could add in to little snippets that you could add in to little snippets that you could add in to make the cursor change and all of these make the cursor change and all of these make the cursor change and all of these different uh different things and then different uh different things and then different uh different things and then it just sort of escalated from there.

it just sort of escalated from there. it just sort of escalated from there. Then you upgrade to Dreamweaver and now Then you upgrade to Dreamweaver and now Then you upgrade to Dreamweaver and now you're coding and then you're like well you're coding and then you're like well you're coding and then you're like well how do you make the pages dynamically? how do you make the pages dynamically? how do you make the pages dynamically? you learn PHP and then for me I you learn PHP and then for me I you learn PHP and then for me I exhausted the internet on Danish exhausted the internet on Danish exhausted the internet on Danish language programming advice. language programming advice. language programming advice. >> Mhm. >> Mhm. >> Mhm. >> Um and I was I was around 11 or 12 >> Um and I was I was around 11 or 12 >> Um and I was I was around 11 or 12 and so I just you know went and got and so I just you know went and got and so I just you know went and got addicted to World of Warcraft for four addicted to World of Warcraft for four addicted to World of Warcraft for four years but that gets you really really years but that gets you really really years but that gets you really really good at English. [laughter] good at English. [laughter] good at English. [laughter] So you kind of start hacking get into So you kind of start hacking get into So you kind of start hacking get into deeper. Now the logical step would have deeper. Now the logical step would have deeper. Now the logical step would have been to just you know go to university been to just you know go to university been to just you know go to university and learn properly about this stuff. But and learn properly about this stuff. But and learn properly about this stuff. But that's not what you did did you? I mean that's not what you did did you? I mean that's not what you did did you? I mean I just um I I just um I I just um I I started just I I mean you know then I I started just I I mean you know then I I started just I I mean you know then I learned video games then I learned learned video games then I learned learned video games then I learned English and then you know this like English and then you know this like English and then you know this like massive arsenal of the web. I now it'd massive arsenal of the web. I now it'd massive arsenal of the web. I now it'd be very interesting because the LLMs be very interesting because the LLMs be very interesting because the LLMs would just speak Danish to me and you would just speak Danish to me and you would just speak Danish to me and you could just you wouldn't have hit the could just you wouldn't have hit the could just you wouldn't have hit the wall like I did. Um, so that would have wall like I did. Um, so that would have wall like I did. Um, so that would have been very interesting. Maybe I would been very interesting. Maybe I would been very interesting. Maybe I would have been better at programming. That have been better at programming. That have been better at programming. That would have been nice. And then I Yeah.

would have been nice. And then I Yeah. would have been nice. And then I Yeah. Then I just started picking up jobs and Then I just started picking up jobs and Then I just started picking up jobs and things like that throughout high school. things like that throughout high school. things like that throughout high school. And when I was in high school as well, I And when I was in high school as well, I And when I was in high school as well, I got exposed to this thing called the got exposed to this thing called the got exposed to this thing called the International Olympiad in Informatics. International Olympiad in Informatics. International Olympiad in Informatics. You heard of this thing? You heard of this thing? You heard of this thing? >> Yeah. Um, and I had a I had an internet >> Yeah. Um, and I had a I had an internet >> Yeah. Um, and I had a I had an internet friend and she lived in Australia and friend and she lived in Australia and friend and she lived in Australia and she was on the Australian team and she she was on the Australian team and she she was on the Australian team and she told there's probably something for the told there's probably something for the told there's probably something for the Danish team as well, but I had never I'd Danish team as well, but I had never I'd Danish team as well, but I had never I'd never heard about it before. And so I never heard about it before. And so I never heard about it before. And so I found it on some like little mysterious found it on some like little mysterious found it on some like little mysterious website and then applied and then solved website and then applied and then solved website and then applied and then solved these programming problems that look these programming problems that look these programming problems that look very different from the HTML and PHP very different from the HTML and PHP very different from the HTML and PHP things that I'd solved until things that I'd solved until things that I'd solved until >> were like the algorithmicalish programs. >> were like the algorithmicalish programs. >> were like the algorithmicalish programs. >> Exactly. It's sort of like this is not >> Exactly. It's sort of like this is not >> Exactly. It's sort of like this is not actually the kind of problem you would actually the kind of problem you would actually the kind of problem you would see there but I think it illustrates see there but I think it illustrates see there but I think it illustrates well the kind of problem that you might well the kind of problem that you might well the kind of problem that you might get right is you could imagine something get right is you could imagine something get right is you could imagine something like okay here's like n trucks here's m like okay here's like n trucks here's m like okay here's like n trucks here's m packages the m packages have these packages the m packages have these packages the m packages have these dimensions dimensions dimensions give me which trucks which packages give me which trucks which packages give me which trucks which packages should be in right and then do something should be in right and then do something should be in right and then do something optimal like that's an npmplete problem optimal like that's an npmplete problem optimal like that's an npmplete problem you can't solve that but you could you can't solve that but you could you can't solve that but you could compete with everyone else in the compete with everyone else in the compete with everyone else in the competition of doing the best thing so competition of doing the best thing so competition of doing the best thing so it's these kinds of problems right it's these kinds of problems right it's these kinds of problems right >> um and so I started doing that. In high >> um and so I started doing that. In high >> um and so I started doing that. In high school, I was working um I was working school, I was working um I was working school, I was working um I was working as well um for a startup. Um and then I as well um for a startup. Um and then I as well um for a startup. Um and then I just Shopify found me while I was still just Shopify found me while I was still just Shopify found me while I was still in high school.

in high school. in high school. >> And and the whole like Shopify found me, >> And and the whole like Shopify found me, >> And and the whole like Shopify found me, was it through your open source was it through your open source was it through your open source contributions? Was it was it something contributions? Was it was it something contributions? Was it was it something else? It was because I had written an else? It was because I had written an else? It was because I had written an article where I had I had I dropped my article where I had I had I dropped my article where I had I had I dropped my iPhone and it was you know the iPhones iPhone and it was you know the iPhones iPhone and it was you know the iPhones are a lot like there used to be a time are a lot like there used to be a time are a lot like there used to be a time right where you drop your iPhone and you right where you drop your iPhone and you right where you drop your iPhone and you just knew it was over for the screen. just knew it was over for the screen. just knew it was over for the screen. >> It doesn't really happen as much anymore >> It doesn't really happen as much anymore >> It doesn't really happen as much anymore like the screens have gotten a lot like the screens have gotten a lot like the screens have gotten a lot better but back then it was like yeah better but back then it was like yeah better but back then it was like yeah one drop and it was dead and it just one drop and it was dead and it just one drop and it was dead and it just couldn't use it anymore. And so I went couldn't use it anymore. And so I went couldn't use it anymore. And so I went back to one of these old Nokia brick back to one of these old Nokia brick back to one of these old Nokia brick phones. And this is back in 2013. And phones. And this is back in 2013. And phones. And this is back in 2013. And people hadn't really realized all the people hadn't really realized all the people hadn't really realized all the pernicious effects of smartphones at the pernicious effects of smartphones at the pernicious effects of smartphones at the time. And so I wrote this article about time. And so I wrote this article about time. And so I wrote this article about how oh my god I'm like calling people how oh my god I'm like calling people how oh my god I'm like calling people and I have my sense of direction back. and I have my sense of direction back. and I have my sense of direction back. Um and I wrote an article about it. And Um and I wrote an article about it. And Um and I wrote an article about it. And this article it went on hacker news this article it went on hacker news this article it went on hacker news briefly and it um New York Times decided briefly and it um New York Times decided briefly and it um New York Times decided to feature it. to feature it. to feature it. >> No way. >> No way. >> No way. >> Yeah. And so a lot of traffic was driven >> Yeah. And so a lot of traffic was driven >> Yeah. And so a lot of traffic was driven to it and then some astute Shopify to it and then some astute Shopify to it and then some astute Shopify recruiter put it all together and um and recruiter put it all together and um and recruiter put it all together and um and I had a call with them and then I don't I had a call with them and then I don't I had a call with them and then I don't think they realized that I was still in think they realized that I was still in think they realized that I was still in high school but um but I had a great high school but um but I had a great high school but um but I had a great call with them. They invited me on site call with them. They invited me on site call with them. They invited me on site to Ottawa, Canada. Um I had no idea what to Ottawa, Canada. Um I had no idea what to Ottawa, Canada. Um I had no idea what Ottawa Canada is. I think the email says Ottawa Canada is. I think the email says Ottawa Canada is. I think the email says something like what's an Ottawa? I had something like what's an Ottawa? I had something like what's an Ottawa? I had no idea. Um, and so I went there and it no idea. Um, and so I went there and it no idea. Um, and so I went there and it was just like walked into the building was just like walked into the building was just like walked into the building and it was just a just felt right. Um, and it was just a just felt right. Um, and it was just a just felt right. Um, and so I I I interviewed with them and and so I I I interviewed with them and and so I I I interviewed with them and then said, "Well, I got to finish high then said, "Well, I got to finish high then said, "Well, I got to finish high school first and then uh and then I

school first and then uh and then I school first and then uh and then I moved to Canada uh to to to work at moved to Canada uh to to to work at moved to Canada uh to to to work at Shopify." Yeah. In 2013. Shopify." Yeah. In 2013. Shopify." Yeah. In 2013. >> Yeah. I think that's that's a like legit >> Yeah. I think that's that's a like legit >> Yeah. I think that's that's a like legit excuse for like not even worrying about excuse for like not even worrying about excuse for like not even worrying about college and and university. college and and university. college and and university. >> But I did it crossed your mind. >> But I did it crossed your mind. >> But I did it crossed your mind. >> It did. I thought I was going I thought >> It did. I thought I was going I thought >> It did. I thought I was going I thought I was doing a gap year. I thought I was I was doing a gap year. I thought I was I was doing a gap year. I thought I was like, "Okay, I'm gonna go work at like, "Okay, I'm gonna go work at like, "Okay, I'm gonna go work at Shopify for a year and then I'll Shopify for a year and then I'll Shopify for a year and then I'll probably go back and do but I would just probably go back and do but I would just probably go back and do but I would just I was very insecure at the time about I was very insecure at the time about I was very insecure at the time about the fact that I hadn't studied computer the fact that I hadn't studied computer the fact that I hadn't studied computer science and my only exposure had been science and my only exposure had been science and my only exposure had been all the II competitions which is a all the II competitions which is a all the II competitions which is a pretty good crash course in a lot of pretty good crash course in a lot of pretty good crash course in a lot of computer science and if nothing else it computer science and if nothing else it computer science and if nothing else it had really taught me that you can just had really taught me that you can just had really taught me that you can just sit down and read a paper and just sit down and read a paper and just sit down and read a paper and just figure it out if you spend enough time figure it out if you spend enough time figure it out if you spend enough time on it." So I did I did that repeatedly on it." So I did I did that repeatedly on it." So I did I did that repeatedly and in my first year at Shopify I just and in my first year at Shopify I just and in my first year at Shopify I just every time I heard something that I every time I heard something that I every time I heard something that I didn't know what was I noted it down on didn't know what was I noted it down on didn't know what was I noted it down on a piece of paper and then I went home a piece of paper and then I went home a piece of paper and then I went home and then that evening I would just read and then that evening I would just read and then that evening I would just read about it because I felt insecure that about it because I felt insecure that about it because I felt insecure that like well if someone mentions mentions like well if someone mentions mentions like well if someone mentions mentions like TCP surely they know exactly what's like TCP surely they know exactly what's like TCP surely they know exactly what's in the three-way handshake and how TLS in the three-way handshake and how TLS in the three-way handshake and how TLS is like layered on top and they've is like layered on top and they've is like layered on top and they've looked at Wireshark and all of that. I looked at Wireshark and all of that. I looked at Wireshark and all of that. I don't think that's true but that's what don't think that's true but that's what don't think that's true but that's what I thought.

I thought. I thought. >> So I went and did that for everything >> So I went and did that for everything >> So I went and did that for everything that I encountered. Um, so that was a that I encountered. Um, so that was a that I encountered. Um, so that was a really good crash course and then very really good crash course and then very really good crash course and then very quickly it became clear that well I just quickly it became clear that well I just quickly it became clear that well I just want to continue doing this. I don't want to continue doing this. I don't want to continue doing this. I don't want to go go somewhere else and then want to go go somewhere else and then want to go go somewhere else and then come back to this because I felt like come back to this because I felt like come back to this because I felt like I'd already found what I wanted to do. I'd already found what I wanted to do. I'd already found what I wanted to do. So it sounds sounds like it was a pretty So it sounds sounds like it was a pretty So it sounds sounds like it was a pretty good combination of like you just having good combination of like you just having good combination of like you just having this like very natural insecurity like this like very natural insecurity like this like very natural insecurity like you know you're young, you know, you you know you're young, you know, you you know you're young, you know, you don't have the education that everyone don't have the education that everyone don't have the education that everyone else has and inside a company that's else has and inside a company that's else has and inside a company that's just doing pretty like cutting edge just doing pretty like cutting edge just doing pretty like cutting edge stuff even at the time and even even to stuff even at the time and even even to stuff even at the time and even even to today, right? Like they're they're today, right? Like they're they're today, right? Like they're they're leading. So you just kept self-seing leading. So you just kept self-seing leading. So you just kept self-seing yourself like just catching up and go yourself like just catching up and go yourself like just catching up and go and then do I understand that you just and then do I understand that you just and then do I understand that you just went deep in every concept that you went deep in every concept that you went deep in every concept that you understood. You didn't like just like understood. You didn't like just like understood. You didn't like just like try to understand a surface level but try to understand a surface level but try to understand a surface level but like go as deep as you can search on the like go as deep as you can search on the like go as deep as you can search on the internet buy books whatever that is. I internet buy books whatever that is. I internet buy books whatever that is. I think it was just that think it was just that think it was just that I just wanted to know keep learning how I just wanted to know keep learning how I just wanted to know keep learning how computers work and I think that this is computers work and I think that this is computers work and I think that this is something that I now look for when we something that I now look for when we something that I now look for when we interview engineers is that you just you interview engineers is that you just you interview engineers is that you just you can't help yourself but trying to peel can't help yourself but trying to peel can't help yourself but trying to peel back the layers and for me that ended up back the layers and for me that ended up back the layers and for me that ended up with the infrastructure layer that was with the infrastructure layer that was with the infrastructure layer that was you know the people closest to the metal you know the people closest to the metal you know the people closest to the metal at at Shopify and I would just always at at Shopify and I would just always at at Shopify and I would just always sit next to them at lunch because I was sit next to them at lunch because I was sit next to them at lunch because I was working on the on the product side but I working on the on the product side but I working on the on the product side but I just I couldn't help myself. I was so I just I couldn't help myself. I was so I just I couldn't help myself. I was so I just wanted to learn what it was when just wanted to learn what it was when just wanted to learn what it was when they were talking about a reverse proxy.

they were talking about a reverse proxy. they were talking about a reverse proxy. I'm like, why is it reverse? I I still I'm like, why is it reverse? I I still I'm like, why is it reverse? I I still can't answer that. [snorts] [snorts] I I I mean, okay, [laughter] I I I mean, okay, [laughter] I I I mean, okay, [laughter] you know, you know, you know, well, what's in reverse? Because it's a well, what's in reverse? Because it's a well, what's in reverse? Because it's a proxy, right? proxy, right? proxy, right? I don't I don't know. I don't know. It's I don't I don't know. I don't know. It's I don't I don't know. I don't know. It's like an inverted index. Like what's like an inverted index. Like what's like an inverted index. Like what's inverted? It's like it's a terrible inverted? It's like it's a terrible inverted? It's like it's a terrible name. Anyway, yeah, name. Anyway, yeah, name. Anyway, yeah, I I mean it's still better when when you I I mean it's still better when when you I I mean it's still better when when you get the NAT nat tables, the lookups, get the NAT nat tables, the lookups, get the NAT nat tables, the lookups, some of those things like some of that. some of those things like some of that. some of those things like some of that. But yeah, I hear you there. There's some But yeah, I hear you there. There's some But yeah, I hear you there. There's some like weird names with this, but at at like weird names with this, but at at like weird names with this, but at at Shopify, Shopify, Shopify, what were some of the kind of like hard what were some of the kind of like hard what were some of the kind of like hard engineuring challenges that you engineuring challenges that you engineuring challenges that you engineering challenges, outages, engineering challenges, outages, engineering challenges, outages, like like learnings that kind of defined like like learnings that kind of defined like like learnings that kind of defined you that were really also fun at the you that were really also fun at the you that were really also fun at the time or interesting to learn, but it time or interesting to learn, but it time or interesting to learn, but it would have been hard to get it would have been hard to get it would have been hard to get it elsewhere. Yeah. So I think it was, you elsewhere. Yeah. So I think it was, you elsewhere. Yeah. So I think it was, you know, in the 2010s there's like a bunch know, in the 2010s there's like a bunch know, in the 2010s there's like a bunch of SAS companies that that scale really of SAS companies that that scale really of SAS companies that that scale really quickly and I felt so fortunate to have quickly and I felt so fortunate to have quickly and I felt so fortunate to have a front row seat to that and so I ended a front row seat to that and so I ended a front row seat to that and so I ended up on the infrastructure team and this up on the infrastructure team and this up on the infrastructure team and this was back in you know 134 and uh Docker was back in you know 134 and uh Docker was back in you know 134 and uh Docker was coming out and so we were was coming out and so we were was coming out and so we were containerizing everything and we were containerizing everything and we were containerizing everything and we were just every single year we had to you just every single year we had to you just every single year we had to you know the growth rates of of of SAS know the growth rates of of of SAS know the growth rates of of of SAS sometimes seems quaint in comparison to sometimes seems quaint in comparison to sometimes seems quaint in comparison to the growth rates of companies today but the growth rates of companies today but the growth rates of companies today but it was a company that was growing at you it was a company that was growing at you it was a company that was growing at you know 120 40% year-over-year. Um, and so know 120 40% year-over-year. Um, and so know 120 40% year-over-year. Um, and so every year we were just preparing for a

every year we were just preparing for a every year we were just preparing for a Black Friday that was going to be a lot Black Friday that was going to be a lot Black Friday that was going to be a lot worse than the last. And this is back in worse than the last. And this is back in worse than the last. And this is back in the day of we're buying physical the day of we're buying physical the day of we're buying physical hardware, right? We have to like place hardware, right? We have to like place hardware, right? We have to like place an order at a particular point in time an order at a particular point in time an order at a particular point in time and do some interpolation based on that. and do some interpolation based on that. and do some interpolation based on that. Um, and the software also had to scale. Um, and the software also had to scale. Um, and the software also had to scale. And when you're scaling most software, a And when you're scaling most software, a And when you're scaling most software, a lot of the application layer problems lot of the application layer problems lot of the application layer problems end up back at the database layer. end up back at the database layer. end up back at the database layer. >> And so I just naturally found myself at >> And so I just naturally found myself at >> And so I just naturally found myself at this layer between Rails and the this layer between Rails and the this layer between Rails and the databases. Shopify didn't at the time at databases. Shopify didn't at the time at databases. Shopify didn't at the time at least contribute many patches to the least contribute many patches to the least contribute many patches to the databases themselves but mostly just databases themselves but mostly just databases themselves but mostly just spent time orchestrating. So we were spent time orchestrating. So we were spent time orchestrating. So we were doing sharding because as um my my dear doing sharding because as um my my dear doing sharding because as um my my dear boss Camilo used to say you can't cache boss Camilo used to say you can't cache boss Camilo used to say you can't cache rights. So there's a fundamental point rights. So there's a fundamental point rights. So there's a fundamental point where you you just you have to move where you you just you have to move where you you just you have to move beyond a single shard. Um so I wasn't I beyond a single shard. Um so I wasn't I beyond a single shard. Um so I wasn't I joined around the time and they did the joined around the time and they did the joined around the time and they did the sharding and they did it I think they sharding and they did it I think they sharding and they did it I think they did the cut over a week before Black did the cut over a week before Black did the cut over a week before Black Friday which is mindblowing. Friday which is mindblowing. Friday which is mindblowing. uh and very but it worked and then the uh and very but it worked and then the uh and very but it worked and then the the subsequent years we worked on things the subsequent years we worked on things the subsequent years we worked on things like going into multiple data centers.

like going into multiple data centers. like going into multiple data centers. We also had this big mysterious reddish We also had this big mysterious reddish We also had this big mysterious reddish server that was like you know 128 GB of server that was like you know 128 GB of server that was like you know 128 GB of RAM which was a lot at the time. Today RAM which was a lot at the time. Today RAM which was a lot at the time. Today it's not that much and no one really it's not that much and no one really it's not that much and no one really knew what was in it and then it went knew what was in it and then it went knew what was in it and then it went down one day and people were like well down one day and people were like well down one day and people were like well that's super terrifying. Um because that's super terrifying. Um because that's super terrifying. Um because people had just been treating it as this people had just been treating it as this people had just been treating it as this KV store. Um and so we started splitting KV store. Um and so we started splitting KV store. Um and so we started splitting it out. We did all this stuff around it out. We did all this stuff around it out. We did all this stuff around making sure that if you if you if you go making sure that if you if you if you go making sure that if you if you if you go visit a Shopify store and the thing that visit a Shopify store and the thing that visit a Shopify store and the thing that stores your sessions is down, the right stores your sessions is down, the right stores your sessions is down, the right behavior is not just for the entire the behavior is not just for the entire the behavior is not just for the entire the of everything to be down. But that's of everything to be down. But that's of everything to be down. But that's kind of the default failure mode, right? kind of the default failure mode, right? kind of the default failure mode, right? You're not going to rescue all of that. You're not going to rescue all of that. You're not going to rescue all of that. Um unless you're in a programming Um unless you're in a programming Um unless you're in a programming language that really forces that language that really forces that language that really forces that decision. So we did things like um build decision. So we did things like um build decision. So we did things like um build this matrix out of okay well this this matrix out of okay well this this matrix out of okay well this service when this component is down service when this component is down service when this component is down should act this way. Um and I found should act this way. Um and I found should act this way. Um and I found myself writing the test suite for a myself writing the test suite for a myself writing the test suite for a bunch of that. And then I was like okay bunch of that. And then I was like okay bunch of that. And then I was like okay well we can't just mock all of this. And well we can't just mock all of this. And well we can't just mock all of this. And so um I came up with this idea at the so um I came up with this idea at the so um I came up with this idea at the time of like oh what we're going to do time of like oh what we're going to do time of like oh what we're going to do is we're just going to um shell out to is we're just going to um shell out to is we're just going to um shell out to GDB and then into the process and then GDB and then into the process and then GDB and then into the process and then close the file descriptor to the close the file descriptor to the close the file descriptor to the database to simulate the through the database to simulate the through the database to simulate the through the entire layer that the database fails.

entire layer that the database fails. entire layer that the database fails. That was a little crazy and we never That was a little crazy and we never That was a little crazy and we never shipped that on CI but it did uncover a shipped that on CI but it did uncover a shipped that on CI but it did uncover a massive amount of issues in Rails that' massive amount of issues in Rails that' massive amount of issues in Rails that' be upstream and things like that of be upstream and things like that of be upstream and things like that of around just like handling failures at around just like handling failures at around just like handling failures at the connection layer. So then I moved on the connection layer. So then I moved on the connection layer. So then I moved on to create this proxy called Toxyroxy and to create this proxy called Toxyroxy and to create this proxy called Toxyroxy and >> have you heard of this before? >> have you heard of this before? >> have you heard of this before? >> No. No. >> No. No. >> No. No. >> Yeah. Toxyroxy is it's just like a layer >> Yeah. Toxyroxy is it's just like a layer >> Yeah. Toxyroxy is it's just like a layer 7 proxy that sits in between um you and 7 proxy that sits in between um you and 7 proxy that sits in between um you and well layer four but in between you and well layer four but in between you and well layer four but in between you and the databases. So you basically have the databases. So you basically have the databases. So you basically have just like this proxy and then my SQL just like this proxy and then my SQL just like this proxy and then my SQL whatever doesn't speak the protocol but whatever doesn't speak the protocol but whatever doesn't speak the protocol but then you can do an API call say take then you can do an API call say take then you can do an API call say take take uh take the database down make it take uh take the database down make it take uh take the database down make it slow um and over time it also added slow um and over time it also added slow um and over time it also added layer 7 things of like do a bunch of layer 7 things of like do a bunch of layer 7 things of like do a bunch of failures this way you're not mocking the failures this way you're not mocking the failures this way you're not mocking the low-level drivers but you're testing the low-level drivers but you're testing the low-level drivers but you're testing the drivers and their failure handling as drivers and their failure handling as drivers and their failure handling as well so then this entire matrix could be well so then this entire matrix could be well so then this entire matrix could be implemented in CI so the basically the implemented in CI so the basically the implemented in CI so the basically the proxy was just like a really thin layer proxy was just like a really thin layer proxy was just like a really thin layer which like was passed through but you which like was passed through but you which like was passed through but you built the functionality to like simulate built the functionality to like simulate built the functionality to like simulate problems with database or things like problems with database or things like problems with database or things like data corruption or whatever you wanted data corruption or whatever you wanted data corruption or whatever you wanted to do. So you could just do it in there to do. So you could just do it in there to do. So you could just do it in there and then you can anything that built on and then you can anything that built on and then you can anything that built on top of it but but then Oh yeah and then top of it but but then Oh yeah and then top of it but but then Oh yeah and then everyone had to like call this proxy or everyone had to like call this proxy or everyone had to like call this proxy or it needed to be on in a layer.

it needed to be on in a layer. it needed to be on in a layer. >> Exactly. So you could do like do like my >> Exactly. So you could do like do like my >> Exactly. So you could do like do like my you know proxy.mmysql you know proxy.mmysql you know proxy.mmysql downdown and then pass it a lambda of downdown and then pass it a lambda of downdown and then pass it a lambda of what you wanted to do like get this page what you wanted to do like get this page what you wanted to do like get this page do a checkout whatever with the sessions do a checkout whatever with the sessions do a checkout whatever with the sessions table down and this just uncovered table down and this just uncovered table down and this just uncovered tens of issues right in the myql driver tens of issues right in the myql driver tens of issues right in the myql driver in the rails like it's just like no one in the rails like it's just like no one in the rails like it's just like no one in the ecosystem had been testing for in the ecosystem had been testing for in the ecosystem had been testing for this and it was very difficult to see this and it was very difficult to see this and it was very difficult to see this in prod right because in myql down this in prod right because in myql down this in prod right because in myql down you're focused on just getting back up you're focused on just getting back up you're focused on just getting back up and not like what could the application and not like what could the application and not like what could the application actually have done. Yeah, it's actually have done. Yeah, it's actually have done. Yeah, it's interesting. Of course, we're going to interesting. Of course, we're going to interesting. Of course, we're going to talk a bit more about databases talk a bit more about databases talk a bit more about databases obviously, but just thinking about how a obviously, but just thinking about how a obviously, but just thinking about how a lot of the problems or some of the most lot of the problems or some of the most lot of the problems or some of the most gnarly problems in large systems are gnarly problems in large systems are gnarly problems in large systems are always to do with state and I never always to do with state and I never always to do with state and I never connected until now that I mean state is connected until now that I mean state is connected until now that I mean state is usually there's a database. If there's usually there's a database. If there's usually there's a database. If there's no database, if you have stateless no database, if you have stateless no database, if you have stateless services, you know, I mean, you still services, you know, I mean, you still services, you know, I mean, you still have problems, you have nodes going have problems, you have nodes going have problems, you have nodes going down, you have, I don't know, down, you have, I don't know, down, you have, I don't know, corruption, whatever, but it's usually corruption, whatever, but it's usually corruption, whatever, but it's usually like more isolated. But basically, like like more isolated. But basically, like like more isolated. But basically, like if we have state, we typically have if we have state, we typically have if we have state, we typically have databases. is if we have databases and databases. is if we have databases and databases. is if we have databases and if you can simulate these problems if you can simulate these problems if you can simulate these problems suddenly you can I mean you you can like suddenly you can I mean you you can like suddenly you can I mean you you can like predict a lot of things but the problem predict a lot of things but the problem predict a lot of things but the problem with state often time is it's really with state often time is it's really with state often time is it's really hard to simulate problems happening hard to simulate problems happening hard to simulate problems happening ahead of time unless when they happen so ahead of time unless when they happen so ahead of time unless when they happen so did you it sounds like you have pretty did you it sounds like you have pretty did you it sounds like you have pretty good success with good success with good success with >> yeah I think to my knowledge it's still >> yeah I think to my knowledge it's still >> yeah I think to my knowledge it's still um running in like the CI system of um running in like the CI system of um running in like the CI system of Shopify today I don't know if anyone in Shopify today I don't know if anyone in Shopify today I don't know if anyone in the crowd is from Shopify but I'm pretty the crowd is from Shopify but I'm pretty the crowd is from Shopify but I'm pretty sure that it still does um and so we sure that it still does um and so we sure that it still does um and so we wrote all these tests against it to wrote all these tests against it to wrote all these tests against it to implement and all of these different uh implement and all of these different uh implement and all of these different uh different failure conditions and it just different failure conditions and it just different failure conditions and it just yeah it was it was it worked out great.

yeah it was it was it worked out great. yeah it was it was it worked out great. So you spent eight years in total at at So you spent eight years in total at at So you spent eight years in total at at Shopify. So like starting from like all Shopify. So like starting from like all Shopify. So like starting from like all right just a gap year it just went on a right just a gap year it just went on a right just a gap year it just went on a year a year and another year. Um at what year a year and another year. Um at what year a year and another year. Um at what point did you think about leaving and point did you think about leaving and point did you think about leaving and why and what was your kind of decision why and what was your kind of decision why and what was your kind of decision framework? It sounds like you or you framework? It sounds like you or you framework? It sounds like you or you were like on Epic right now even today were like on Epic right now even today were like on Epic right now even today Shopify it's doing wonderful. It's Shopify it's doing wonderful. It's Shopify it's doing wonderful. It's probably doing even way better than like probably doing even way better than like probably doing even way better than like like you know that growth kind of kept like you know that growth kind of kept like you know that growth kind of kept on. So I'm sure there would have been an on. So I'm sure there would have been an on. So I'm sure there would have been an argument to stay and you know stay on argument to stay and you know stay on argument to stay and you know stay on their August ship. their August ship. their August ship. >> Yeah. So I I spent I spent eight years >> Yeah. So I I spent I spent eight years >> Yeah. So I I spent I spent eight years there from 13 to to 21. Um and I I think there from 13 to to 21. Um and I I think there from 13 to to 21. Um and I I think there just came a point where there just came a point where there just came a point where I wanted to see something different I wanted to see something different I wanted to see something different again. I've been inside of Shopify since again. I've been inside of Shopify since again. I've been inside of Shopify since I was 18 years old, right? I'd been seen I was 18 years old, right? I'd been seen I was 18 years old, right? I'd been seen one other startup in high school. I was one other startup in high school. I was one other startup in high school. I was like if I want to learn more about like if I want to learn more about like if I want to learn more about computers and learn faster, it might be computers and learn faster, it might be computers and learn faster, it might be time to inject some novelty into this time to inject some novelty into this time to inject some novelty into this function. Um, and so I I left in in in function. Um, and so I I left in in in function. Um, and so I I left in in in 21 and I'd worked on so many different 21 and I'd worked on so many different 21 and I'd worked on so many different parts of the infrastructure like parts of the infrastructure like parts of the infrastructure like caching. Um, me and Justine, who's now caching. Um, me and Justine, who's now caching. Um, me and Justine, who's now my co-founder, we wrote the entire my co-founder, we wrote the entire my co-founder, we wrote the entire storefront um, storefront for Shopify, storefront um, storefront for Shopify, storefront um, storefront for Shopify, um, which powered almost 100% of traffic um, which powered almost 100% of traffic um, which powered almost 100% of traffic 18 months after we embarked on it. Um, 18 months after we embarked on it. Um, 18 months after we embarked on it. Um, we've worked on running Shopify in we've worked on running Shopify in we've worked on running Shopify in multiple data centers. We've worked on multiple data centers. We've worked on multiple data centers. We've worked on so many database scaling projects like so many database scaling projects like so many database scaling projects like caching, all of these different things, caching, all of these different things, caching, all of these different things, right? Um, a lot of the a lot of the right? Um, a lot of the a lot of the right? Um, a lot of the a lot of the scalability came from the Kardashians scalability came from the Kardashians scalability came from the Kardashians launching lots of products on on launching lots of products on on launching lots of products on on Shopify, which would force a lot of Shopify, which would force a lot of Shopify, which would force a lot of traffic. Um, but that's that's traffic. Um, but that's that's traffic. Um, but that's that's eventually how I left. And so when I eventually how I left. And so when I eventually how I left. And so when I left, I didn't really know what I wanted

left, I didn't really know what I wanted left, I didn't really know what I wanted to do. And so I one of the projects I to do. And so I one of the projects I to do. And so I one of the projects I had while I was at Shopify was this had while I was at Shopify was this had while I was at Shopify was this napkin math project. Have you seen this napkin math project. Have you seen this napkin math project. Have you seen this >> napkin math? No. >> napkin math? No. >> napkin math? No. >> No. Um, so napkin math was essentially >> No. Um, so napkin math was essentially >> No. Um, so napkin math was essentially just this table that I maintain on just this table that I maintain on just this table that I maintain on GitHub of how much bandwidth can you GitHub of how much bandwidth can you GitHub of how much bandwidth can you drive to DRAMM, what does a roundtrip to drive to DRAMM, what does a roundtrip to drive to DRAMM, what does a roundtrip to S3 cost, and how long does it take, how S3 cost, and how long does it take, how S3 cost, and how long does it take, how much bandwidth can you drive to an NVME much bandwidth can you drive to an NVME much bandwidth can you drive to an NVME SSD, how much bandwidth can you drive to SSD, how much bandwidth can you drive to SSD, how much bandwidth can you drive to an EBS volume? Just a collection of an EBS volume? Just a collection of an EBS volume? Just a collection of probably there's probably like 50 of probably there's probably like 50 of probably there's probably like 50 of these numbers and then a RS script that these numbers and then a RS script that these numbers and then a RS script that generates them all. um what all these generates them all. um what all these generates them all. um what all these things cost? What do you like? What does things cost? What do you like? What does things cost? What do you like? What does a gigabyte of memory cost? $2. What does a gigabyte of memory cost? $2. What does a gigabyte of memory cost? $2. What does a gigabyte of S3 cost? 2 cents. What a gigabyte of S3 cost? 2 cents. What a gigabyte of S3 cost? 2 cents. What does a gigabyte of um this cost? 10 does a gigabyte of um this cost? 10 does a gigabyte of um this cost? 10 cents, right? What does it cost on spot? cents, right? What does it cost on spot? cents, right? What does it cost on spot? What does it cost on a three-year What does it cost on a three-year What does it cost on a three-year commit? Like I had just have a massive commit? Like I had just have a massive commit? Like I had just have a massive table and then create flash cards for table and then create flash cards for table and then create flash cards for almost every single cell. So I know all almost every single cell. So I know all almost every single cell. So I know all these numbers. And this was a project I these numbers. And this was a project I these numbers. And this was a project I started taking on at Shopify because I started taking on at Shopify because I started taking on at Shopify because I found myself in um this role a lot where found myself in um this role a lot where found myself in um this role a lot where I would go in and review a project, I would go in and review a project, I would go in and review a project, right? So some product team would be right? So some product team would be right? So some product team would be like okay we got to do we got to build like okay we got to do we got to build like okay we got to do we got to build this thing so we got to build this this thing so we got to build this this thing so we got to build this infrastructure to support the feature infrastructure to support the feature infrastructure to support the feature and a lot of the times they would say and a lot of the times they would say and a lot of the times they would say okay well we've gone and benchmarked it okay well we've gone and benchmarked it okay well we've gone and benchmarked it on database A on database A on database A but the benchmarks are not very good so but the benchmarks are not very good so but the benchmarks are not very good so we're going to go with database B we're going to go with database B we're going to go with database B and I hate benchmarks so much because and I hate benchmarks so much because and I hate benchmarks so much because that's not a satisfying answer to me to that's not a satisfying answer to me to that's not a satisfying answer to me to me it's like

me it's like me it's like this does not jive maybe my intuition this does not jive maybe my intuition this does not jive maybe my intuition database A that you're saying takes 10 database A that you're saying takes 10 database A that you're saying takes 10 seconds to do this seconds to do this seconds to do this should take 10 milliseconds if you do should take 10 milliseconds if you do should take 10 milliseconds if you do the napkin math right if it's a search the napkin math right if it's a search the napkin math right if it's a search query right it's like okay you're query right it's like okay you're query right it's like okay you're searching for three terms there each searching for three terms there each searching for three terms there each term has this many documents that match term has this many documents that match term has this many documents that match it that's this many megabytes we inter it that's this many megabytes we inter it that's this many megabytes we inter intersect these many this many lists intersect these many this many lists intersect these many this many lists you have DRAM bandwidth on multiple you have DRAM bandwidth on multiple you have DRAM bandwidth on multiple cores of 100 gigabytes per second this cores of 100 gigabytes per second this cores of 100 gigabytes per second this should take 10 millisecond you tell me should take 10 millisecond you tell me should take 10 millisecond you tell me the benchmark takes 10 the benchmark takes 10 the benchmark takes 10 one of us is wrong. Either there's a gap one of us is wrong. Either there's a gap one of us is wrong. Either there's a gap in my understanding, which is very in my understanding, which is very in my understanding, which is very likely, or you would benchmark the wrong likely, or you would benchmark the wrong likely, or you would benchmark the wrong thing. And in some ways, some reasons, thing. And in some ways, some reasons, thing. And in some ways, some reasons, right, it's like, okay, you've done a right, it's like, okay, you've done a right, it's like, okay, you've done a benchmark. You don't didn't realize that benchmark. You don't didn't realize that benchmark. You don't didn't realize that your benchmark is doing a distributed your benchmark is doing a distributed your benchmark is doing a distributed query across a 100 different nodes. And query across a 100 different nodes. And query across a 100 different nodes. And so, of course, the P99 is going to be so, of course, the P99 is going to be so, of course, the P99 is going to be really, really high, right? Unless really, really high, right? Unless really, really high, right? Unless you've cut that off or or made some you've cut that off or or made some you've cut that off or or made some different set of trade-offs. So I just different set of trade-offs. So I just different set of trade-offs. So I just found myself in these discussions found myself in these discussions found myself in these discussions repeatedly where people were making repeatedly where people were making repeatedly where people were making infrastructure decisions based on poor infrastructure decisions based on poor infrastructure decisions based on poor benchmarks. And so I needed some I I benchmarks. And so I needed some I I benchmarks. And so I needed some I I needed some ammo to go in and just be needed some ammo to go in and just be needed some ammo to go in and just be like okay we can just do the calculation like okay we can just do the calculation like okay we can just do the calculation right here and then. Um because I was right here and then. Um because I was right here and then. Um because I was always doing these like little demos or always doing these like little demos or always doing these like little demos or like writing little prototype scripts to like writing little prototype scripts to like writing little prototype scripts to to demonstrate this. But it was just I to demonstrate this. But it was just I to demonstrate this. But it was just I just the argument of here's how a beach just the argument of here's how a beach just the argument of here's how a beach tree works. This is how many pages we tree works. This is how many pages we tree works. This is how many pages we have to visit. This is what a random SSD have to visit. This is what a random SSD have to visit. This is what a random SSD read takes. It takes one millisecond.

read takes. It takes one millisecond. read takes. It takes one millisecond. you have to visit a thousand blah blah you have to visit a thousand blah blah you have to visit a thousand blah blah blah blah blah and then present it back blah blah blah and then present it back blah blah blah and then present it back and see if this is the difference to and see if this is the difference to and see if this is the difference to your query. Well, like is the query plan your query. Well, like is the query plan your query. Well, like is the query plan correct? Like is there a bug in my SQL? correct? Like is there a bug in my SQL? correct? Like is there a bug in my SQL? Do we have bad discs? Like what's the Do we have bad discs? Like what's the Do we have bad discs? Like what's the discrepancy here? And I just got caught discrepancy here? And I just got caught discrepancy here? And I just got caught with that bug. And so after I left Sha, with that bug. And so after I left Sha, with that bug. And so after I left Sha, I was just writing a lot of articles I was just writing a lot of articles I was just writing a lot of articles about this. I was just like, well, how about this. I was just like, well, how about this. I was just like, well, how long does should this query take? And long does should this query take? And long does should this query take? And then I one hypothesis I had at some then I one hypothesis I had at some then I one hypothesis I had at some point is like, okay, well, how many point is like, okay, well, how many point is like, okay, well, how many writes per second can MySQL do? Well, writes per second can MySQL do? Well, writes per second can MySQL do? Well, shouldn't the amount of writes per shouldn't the amount of writes per shouldn't the amount of writes per second that MySQL do equal the amount of second that MySQL do equal the amount of second that MySQL do equal the amount of f-syncs that you can do per second? That f-syncs that you can do per second? That f-syncs that you can do per second? That sort of makes sense, right? Every time sort of makes sense, right? Every time sort of makes sense, right? Every time you do a write, you f-sync to persist to you do a write, you f-sync to persist to you do a write, you f-sync to persist to disk. So, how many f-syncs can you do disk. So, how many f-syncs can you do disk. So, how many f-syncs can you do per second? Well, an f-sync takes one per second? Well, an f-sync takes one per second? Well, an f-sync takes one millisecond. So, you do a thousand millisecond. So, you do a thousand millisecond. So, you do a thousand writes per second. That well, that writes per second. That well, that writes per second. That well, that doesn't really match up. Like, feel like doesn't really match up. Like, feel like doesn't really match up. Like, feel like a database can do more than,000 rightes a database can do more than,000 rightes a database can do more than,000 rightes per second. Why can it do that? So, that per second. Why can it do that? So, that per second. Why can it do that? So, that was one of those things where I tested was one of those things where I tested was one of those things where I tested and it's like, okay, well, my SQL on a and it's like, okay, well, my SQL on a and it's like, okay, well, my SQL on a little dinky box could do 10,000 writes little dinky box could do 10,000 writes little dinky box could do 10,000 writes per second. Well, how is that possible? per second. Well, how is that possible? per second. Well, how is that possible? Mhm. Mhm. Mhm. >> And now you would just ask how is it >> And now you would just ask how is it >> And now you would just ask how is it possible? possible? possible? >> Because you batch. So an f-sync happens >> Because you batch. So an f-sync happens >> Because you batch. So an f-sync happens on usually a 4K.

on usually a 4K. on usually a 4K. >> Yeah. >> Yeah. >> Yeah. >> Right. But it's like that's not >> Right. But it's like that's not >> Right. But it's like that's not intuitive. Like it's actually I I I got intuitive. Like it's actually I I I got intuitive. Like it's actually I I I got caught. It was like just like you know caught. It was like just like you know caught. It was like just like you know probably some like 24-hour period where probably some like 24-hour period where probably some like 24-hour period where I just got obsessed with this question I just got obsessed with this question I just got obsessed with this question where like you're writing like the BPF where like you're writing like the BPF where like you're writing like the BPF traces and all of that to do all of traces and all of that to do all of traces and all of that to do all of this. This is like pre-LM so it took this. This is like pre-LM so it took this. This is like pre-LM so it took forever and you and then I found out forever and you and then I found out forever and you and then I found out that oh every f-sync was like much that oh every f-sync was like much that oh every f-sync was like much larger than I would have inferred like larger than I would have inferred like larger than I would have inferred like oh it's batching you go into the code oh it's batching you go into the code oh it's batching you go into the code and you read it and then you found some and you read it and then you found some and you read it and then you found some obscure article by it's always somewhere obscure article by it's always somewhere obscure article by it's always somewhere in like a central German town that's in like a central German town that's in like a central German town that's like written some article about like how like written some article about like how like written some article about like how some intricacy of my SQL works and a some intricacy of my SQL works and a some intricacy of my SQL works and a patch that they did to it's like the patch that they did to it's like the patch that they did to it's like the entire internet runs on small towns in entire internet runs on small towns in entire internet runs on small towns in Bavaria. I'm convinced. Yeah. Bavaria. I'm convinced. Yeah. Bavaria. I'm convinced. Yeah. And then you decided to start And then you decided to start And then you decided to start Turbopuffer. Turbopuffer. Turbopuffer. Yeah. Did h how did you decide? Did you Yeah. Did h how did you decide? Did you Yeah. Did h how did you decide? Did you know what you wanted to build or was it know what you wanted to build or was it know what you wanted to build or was it more like I want to build something more like I want to build something more like I want to build something something databases because you were something databases because you were something databases because you were clearly very into databases. You you've clearly very into databases. You you've clearly very into databases. You you've done an awesome job benchmarking like done an awesome job benchmarking like done an awesome job benchmarking like what is the theoretical like limits? You what is the theoretical like limits? You what is the theoretical like limits? You were very familiar with this probably were very familiar with this probably were very familiar with this probably became you know like world expert in in became you know like world expert in in became you know like world expert in in this niche. And then how this niche. And then how this niche. And then how >> did did you want to go into databases >> did did you want to go into databases >> did did you want to go into databases again?

again? again? >> I think it was >> I think it was >> I think it was there's three things that sort of came there's three things that sort of came there's three things that sort of came to a head. Um the last project that I to a head. Um the last project that I to a head. Um the last project that I worked on at Shopify was search and I worked on at Shopify was search and I worked on at Shopify was search and I didn't have a good time. didn't have a good time. didn't have a good time. >> What what what did you use back there? >> What what what did you use back there? >> What what what did you use back there? Um I don't we don't need to name names Um I don't we don't need to name names Um I don't we don't need to name names of other database companies but it was of other database companies but it was of other database companies but it was uh it was one of the one of the like uh it was one of the one of the like uh it was one of the one of the like traditional search companies that a lot traditional search companies that a lot traditional search companies that a lot of different um companies run and it was of different um companies run and it was of different um companies run and it was just very difficult to get it to do what just very difficult to get it to do what just very difficult to get it to do what I did and I was just like the projects I did and I was just like the projects I did and I was just like the projects that touched that database just I that touched that database just I that touched that database just I couldn't get them to perform at the couldn't get them to perform at the couldn't get them to perform at the napkin math and like there's there's no napkin math and like there's there's no napkin math and like there's there's no query planner and like I couldn't figure query planner and like I couldn't figure query planner and like I couldn't figure out why it wasn't there and sometimes it out why it wasn't there and sometimes it out why it wasn't there and sometimes it tracked and then sometimes it really tracked and then sometimes it really tracked and then sometimes it really didn't track at all and so I tried to didn't track at all and so I tried to didn't track at all and so I tried to learn as much as I could to figure out learn as much as I could to figure out learn as much as I could to figure out and like start reading the source code and like start reading the source code and like start reading the source code of it and I was just I couldn't get it of it and I was just I couldn't get it of it and I was just I couldn't get it to track very often. It was very to track very often. It was very to track very often. It was very difficult to operate and so I just that difficult to operate and so I just that difficult to operate and so I just that was sort of like in the back of my head. was sort of like in the back of my head. was sort of like in the back of my head. I never thought I would touch that I never thought I would touch that I never thought I would touch that again. Then the second ingredient was again. Then the second ingredient was again. Then the second ingredient was the napkin math project the napkin math project the napkin math project because it sort of just gave me a lot of because it sort of just gave me a lot of because it sort of just gave me a lot of facility with all of these napkin math facility with all of these napkin math facility with all of these napkin math numbers of what might be achievable with numbers of what might be achievable with numbers of what might be achievable with the machine if you utilized it perfectly the machine if you utilized it perfectly the machine if you utilized it perfectly >> properly. Yeah. And then the third one >> properly. Yeah. And then the third one >> properly. Yeah. And then the third one was that doing this you know leaving was that doing this you know leaving was that doing this you know leaving Shopify in 21 having spent eight years Shopify in 21 having spent eight years Shopify in 21 having spent eight years there and during that time I did this I there and during that time I did this I there and during that time I did this I called it angel engineering so I like called it angel engineering so I like called it angel engineering so I like joined my friends companies and then I joined my friends companies and then I joined my friends companies and then I just vested equity instead of um instead just vested equity instead of um instead just vested equity instead of um instead of just investing or something like that of just investing or something like that of just investing or something like that and because I wanted to have my fingers and because I wanted to have my fingers and because I wanted to have my fingers in it I wanted to like see what else was in it I wanted to like see what else was in it I wanted to like see what else was out there that's why I left and this out there that's why I left and this out there that's why I left and this problem kept coming up again again and problem kept coming up again again and problem kept coming up again again and again and again right like ChachiBT came

again and again right like ChachiBT came again and again right like ChachiBT came out in 2022 and I was working with with out in 2022 and I was working with with out in 2022 and I was working with with a company then and They wanted to a company then and They wanted to a company then and They wanted to connect a bunch of documents to AI and connect a bunch of documents to AI and connect a bunch of documents to AI and that's when the context windows were that's when the context windows were that's when the context windows were really small. So you had to reach for really small. So you had to reach for really small. So you had to reach for search very quickly. search very quickly. search very quickly. >> So it's like a few kilobytes. >> So it's like a few kilobytes. >> So it's like a few kilobytes. >> It was eight kilobytes or four kilobytes >> It was eight kilobytes or four kilobytes >> It was eight kilobytes or four kilobytes depending on the model. It was very very depending on the model. It was very very depending on the model. It was very very small. So you had to reach for search small. So you had to reach for search small. So you had to reach for search very quickly. Right. And very quickly. Right. And very quickly. Right. And >> so I I I worked with them and I was I >> so I I I worked with them and I was I >> so I I I worked with them and I was I was I created a little recommendation was I created a little recommendation was I created a little recommendation engine and the recommendation engine was engine and the recommendation engine was engine and the recommendation engine was actually quite good. Um, like I s I actually quite good. Um, like I s I actually quite good. Um, like I s I found out that one of the co-founders found out that one of the co-founders found out that one of the co-founders wife was pregnant through the wife was pregnant through the wife was pregnant through the recommendations that I was getting when recommendations that I was getting when recommendations that I was getting when I was running it on his feed. Um, like I was running it on his feed. Um, like I was running it on his feed. Um, like it was it was it it it was it was it it it was it was it it >> weird but >> weird but >> weird but >> it was recommending. Yeah. I mean it was >> it was recommending. Yeah. I mean it was >> it was recommending. Yeah. I mean it was just like you know he was reading about just like you know he was reading about just like you know he was reading about like and I did get permission. I just like and I did get permission. I just like and I did get permission. I just like I don't think anyone expected to be like I don't think anyone expected to be like I don't think anyone expected to be good enough and just like okay it's this good enough and just like okay it's this good enough and just like okay it's this this thing is working and then I ran the this thing is working and then I ran the this thing is working and then I ran the back of the envelope math on what it back of the envelope math on what it back of the envelope math on what it would cost to do this for everyone like would cost to do this for everyone like would cost to do this for everyone like all the users. This is a company called all the users. This is a company called all the users. This is a company called Readwise. So it's like articles that you Readwise. So it's like articles that you Readwise. So it's like articles that you save and then and insert later. And it save and then and insert later. And it save and then and insert later. And it was going to cost 30 grand a month. And was going to cost 30 grand a month. And was going to cost 30 grand a month. And this was a company. It's a bootstrap this was a company. It's a bootstrap this was a company. It's a bootstrap Canadian company. They spend about five Canadian company. They spend about five Canadian company. They spend about five they at the time they were spending they at the time they were spending they at the time they were spending about 5k a month on all the other about 5k a month on all the other about 5k a month on all the other infrastructure combined.

infrastructure combined. infrastructure combined. So it just it didn't the you know So it just it didn't the you know So it just it didn't the you know fundamentally in a company if you're fundamentally in a company if you're fundamentally in a company if you're doing an investment you have have to run doing an investment you have have to run doing an investment you have have to run some gross margin on top of whatever some gross margin on top of whatever some gross margin on top of whatever you're paying right and it just didn't you're paying right and it just didn't you're paying right and it just didn't line up. Um and so we just didn't ship line up. Um and so we just didn't ship line up. Um and so we just didn't ship it. I worked on I you know tuned to it. I worked on I you know tuned to it. I worked on I you know tuned to autovacuum on postcress or something autovacuum on postcress or something autovacuum on postcress or something like that which is a good pastime and like that which is a good pastime and like that which is a good pastime and then you I just couldn't stop thinking then you I just couldn't stop thinking then you I just couldn't stop thinking about why it was so expensive to store about why it was so expensive to store about why it was so expensive to store all of these vectors that we were using all of these vectors that we were using all of these vectors that we were using for the recommendations and I just sat for the recommendations and I just sat for the recommendations and I just sat and did the napkin math one day of like and did the napkin math one day of like and did the napkin math one day of like can we just use it all in S3 and do some can we just use it all in S3 and do some can we just use it all in S3 and do some clustering and then organize the files clustering and then organize the files clustering and then organize the files and just the way and it's like maybe you and just the way and it's like maybe you and just the way and it's like maybe you could build that and then could build that and then could build that and then one day I just kind of said [ __ ] it and one day I just kind of said [ __ ] it and one day I just kind of said [ __ ] it and did it and like sat down and started did it and like sat down and started did it and like sat down and started started to like to write it out. Um, and started to like to write it out. Um, and started to like to write it out. Um, and I spent the summer of of 23 just I spent the summer of of 23 just I spent the summer of of 23 just hammering my head against the wall hammering my head against the wall hammering my head against the wall trying to find an approach where I could trying to find an approach where I could trying to find an approach where I could get the latency that I wanted. Um, get the latency that I wanted. Um, get the latency that I wanted. Um, >> because the problem with S3 is it has >> because the problem with S3 is it has >> because the problem with S3 is it has really good durability, but latency really good durability, but latency really good durability, but latency we're talking hundreds of milliseconds, we're talking hundreds of milliseconds, we're talking hundreds of milliseconds, right? right? right? >> Yes. The P99 on a uh 256 or 512 kilobyte >> Yes. The P99 on a uh 256 or 512 kilobyte >> Yes. The P99 on a uh 256 or 512 kilobyte object on S3 um is around 200 object on S3 um is around 200 object on S3 um is around 200 milliseconds. Um, milliseconds. Um, milliseconds. Um, >> and and you're saying P99 because like >> and and you're saying P99 because like >> and and you're saying P99 because like when you're talking large scale, you when you're talking large scale, you when you're talking large scale, you want to care about the P99, right?

want to care about the P99, right? want to care about the P99, right? >> Yeah. I think when you're >> Yeah. I think when you're >> Yeah. I think when you're >> That's why we're not talking about P50. >> That's why we're not talking about P50. >> That's why we're not talking about P50. >> When you're designing a system, you want >> When you're designing a system, you want >> When you're designing a system, you want to optimize for the P99. And especially to optimize for the P99. And especially to optimize for the P99. And especially because when you're designing a system because when you're designing a system because when you're designing a system on on S3, generally in every roundtrip, on on S3, generally in every roundtrip, on on S3, generally in every roundtrip, you're not doing one request. You're you're not doing one request. You're you're not doing one request. You're often doing lots of requests, right? often doing lots of requests, right? often doing lots of requests, right? >> You're going to hit the P99 real quick. >> You're going to hit the P99 real quick. >> You're going to hit the P99 real quick. >> Exactly. So, it's like if you're >> Exactly. So, it's like if you're >> Exactly. So, it's like if you're navigating a tree on S3, right? It's navigating a tree on S3, right? It's navigating a tree on S3, right? It's like, okay, you get the upper layer of like, okay, you get the upper layer of like, okay, you get the upper layer of the tree 200 milliseconds. You get like the tree 200 milliseconds. You get like the tree 200 milliseconds. You get like another layer of the tree 200 another layer of the tree 200 another layer of the tree 200 milliseconds. You get a bunch of leaves milliseconds. You get a bunch of leaves milliseconds. You get a bunch of leaves of the tree in 200 millonds. So in of the tree in 200 millonds. So in of the tree in 200 millonds. So in aggregate you have like you want to look aggregate you have like you want to look aggregate you have like you want to look at the P99 probably even the P999 to at the P99 probably even the P999 to at the P99 probably even the P999 to design the system properly because you design the system properly because you design the system properly because you will need to minimize the number of will need to minimize the number of will need to minimize the number of round trips that you had to make. So, I round trips that you had to make. So, I round trips that you had to make. So, I just sat and sketched that out um and just sat and sketched that out um and just sat and sketched that out um and tried a bunch of different approaches tried a bunch of different approaches tried a bunch of different approaches and then and then finally in in in July and then and then finally in in in July and then and then finally in in in July of 23, I I I got something end to end of 23, I I I got something end to end of 23, I I I got something end to end that seemed to work and then rewrote it that seemed to work and then rewrote it that seemed to work and then rewrote it probably twice and then released it in probably twice and then released it in probably twice and then released it in in October of of 23 based on um based on in October of of 23 based on um based on in October of of 23 based on um based on just that that summer of of of working just that that summer of of of working just that that summer of of of working through it. And then you kind of you through it. And then you kind of you through it. And then you kind of you built it on on top of S3 because I guess built it on on top of S3 because I guess built it on on top of S3 because I guess durability and and all of and just durability and and all of and just durability and and all of and just really good. How did you make it fast?

really good. How did you make it fast? really good. How did you make it fast? We didn't in the beginning or I didn't We didn't in the beginning or I didn't We didn't in the beginning or I didn't in the beginning. Um it was just me at in the beginning. Um it was just me at in the beginning. Um it was just me at the time and it was really like it was the time and it was really like it was the time and it was really like it was it was it was a project. It was not a it was it was a project. It was not a it was it was a project. It was not a company. It was not company. It was not company. It was not >> it was it was it was to satisfy a >> it was it was it was to satisfy a >> it was it was it was to satisfy a curiosity. It was not I did not set out curiosity. It was not I did not set out curiosity. It was not I did not set out to do this like I'm going to go like to do this like I'm going to go like to do this like I'm going to go like raise $10 million and do like I was like raise $10 million and do like I was like raise $10 million and do like I was like I barely knew what a VC was. Like I was I barely knew what a VC was. Like I was I barely knew what a VC was. Like I was like I just had to do this thing and I like I just had to do this thing and I like I just had to do this thing and I was so focused on doing it. so clear to was so focused on doing it. so clear to was so focused on doing it. so clear to me that if I wasn't going to do it, me that if I wasn't going to do it, me that if I wasn't going to do it, someone else was going to do it and I someone else was going to do it and I someone else was going to do it and I just became fully obsessed that summer just became fully obsessed that summer just became fully obsessed that summer with it. And so the first version was with it. And so the first version was with it. And so the first version was the simplest possible thing. I think I'm the simplest possible thing. I think I'm the simplest possible thing. I think I'm a very pragmatic person like I I didn't a very pragmatic person like I I didn't a very pragmatic person like I I didn't get buried. I barely read any like of get buried. I barely read any like of get buried. I barely read any like of the literature on LSM. I sort of like the literature on LSM. I sort of like the literature on LSM. I sort of like you know read a bunch of it just like you know read a bunch of it just like you know read a bunch of it just like got the basic idea barely implemented got the basic idea barely implemented got the basic idea barely implemented that because that would have taken too that because that would have taken too that because that would have taken too much time. It was the simplest possible much time. It was the simplest possible much time. It was the simplest possible version of what it could be. Like really version of what it could be. Like really version of what it could be. Like really what you have to imagine is that the what you have to imagine is that the what you have to imagine is that the simplest way you could do this is you simplest way you could do this is you simplest way you could do this is you run some clustering algorithm on the run some clustering algorithm on the run some clustering algorithm on the vectors.

vectors. vectors. >> You get the clusters and then you put >> You get the clusters and then you put >> You get the clusters and then you put the clusters in files. The cl the files the clusters in files. The cl the files the clusters in files. The cl the files are called cluster one, cluster two, are called cluster one, cluster two, are called cluster one, cluster two, cluster three and then you have another cluster three and then you have another cluster three and then you have another file called centroidids of the clusters file called centroidids of the clusters file called centroidids of the clusters and then you do the search by and then you do the search by and then you do the search by downloading centroidids looking at the downloading centroidids looking at the downloading centroidids looking at the centrids and then downloading the n centrids and then downloading the n centrids and then downloading the n closest clusters. There was a few closest clusters. There was a few closest clusters. There was a few optimizations around merging some optimizations around merging some optimizations around merging some clusters that were JSON in files and so clusters that were JSON in files and so clusters that were JSON in files and so on just to like control some cost and on just to like control some cost and on just to like control some cost and some performance but that was basically some performance but that was basically some performance but that was basically it and then getting that to scale. That it and then getting that to scale. That it and then getting that to scale. That was the first version. And then how do was the first version. And then how do was the first version. And then how do we make it fast? Well, I didn't even we make it fast? Well, I didn't even we make it fast? Well, I didn't even implement a caching layer. I just put implement a caching layer. I just put implement a caching layer. I just put the reverse proxy in front of S3 with the reverse proxy in front of S3 with the reverse proxy in front of S3 with Engine X and then had it Engine X and then had it Engine X and then had it >> know what a reverse proxy is. >> know what a reverse proxy is. >> know what a reverse proxy is. >> I like I do know what it is. I just >> I like I do know what it is. I just >> I like I do know what it is. I just still don't know what what the reverse still don't know what what the reverse still don't know what what the reverse is about. But anyway, um the reverse the is about. But anyway, um the reverse the is about. But anyway, um the reverse the reverse proxy reverse things. Um the the reverse proxy reverse things. Um the the reverse proxy reverse things. Um the the performance in this case um maybe that's performance in this case um maybe that's performance in this case um maybe that's what it's about by caching right all of what it's about by caching right all of what it's about by caching right all of the all of the S3 objects. It again it the all of the S3 objects. It again it the all of the S3 objects. It again it was the simplest like it's like I'm just was the simplest like it's like I'm just was the simplest like it's like I'm just going to put that in front. I knew how going to put that in front. I knew how going to put that in front. I knew how to configure engine X like I've written to configure engine X like I've written to configure engine X like I've written more enginex Lua than u than a lot of more enginex Lua than u than a lot of more enginex Lua than u than a lot of engineext Lua very good software. um engineext Lua very good software. um engineext Lua very good software. um just had that cache in front and then just had that cache in front and then just had that cache in front and then the way that I would do things like the way that I would do things like the way that I would do things like deleting in the cache was just like deleting in the cache was just like deleting in the cache was just like shell out to XRX and just remove like shell out to XRX and just remove like shell out to XRX and just remove like things in the in the cache and reverse things in the in the cache and reverse things in the in the cache and reverse engineer the directory structure on engineer the directory structure on engineer the directory structure on engine X and that's what we shipped and engine X and that's what we shipped and engine X and that's what we shipped and it was just running on a single server it was just running on a single server it was just running on a single server in a T-Ox instance. I was like okay in a T-Ox instance. I was like okay in a T-Ox instance. I was like okay let's see if anyone gives a [ __ ] Yeah.

let's see if anyone gives a [ __ ] Yeah. let's see if anyone gives a [ __ ] Yeah. So so far I mean this is kind of like So so far I mean this is kind of like So so far I mean this is kind of like cool engineering and like a cool side cool engineering and like a cool side cool engineering and like a cool side project and like a bunch of novel ideas project and like a bunch of novel ideas project and like a bunch of novel ideas and I you know like I think just some and I you know like I think just some and I you know like I think just some hardcore engineering. How did cursor hardcore engineering. How did cursor hardcore engineering. How did cursor come into play? Because like when I come into play? Because like when I come into play? Because like when I learned about Turbopuffer, I was talking learned about Turbopuffer, I was talking learned about Turbopuffer, I was talking with Cursor about like how they built with Cursor about like how they built with Cursor about like how they built their their back end, their database, their their back end, their database, their their back end, their database, how they scaled, and they're telling me how they scaled, and they're telling me how they scaled, and they're telling me all these migrations and they were all these migrations and they were all these migrations and they were telling me like, oh yeah, so we we were telling me like, oh yeah, so we we were telling me like, oh yeah, so we we were on Postgress, but it didn't no they did on Postgress, but it didn't no they did on Postgress, but it didn't no they did something else in Postgress. It didn't something else in Postgress. It didn't something else in Postgress. It didn't even work that well. They went to AWS even work that well. They went to AWS even work that well. They went to AWS Aurora, which is managed service of Aurora, which is managed service of Aurora, which is managed service of Postgress, and it didn't work well, Postgress, and it didn't work well, Postgress, and it didn't work well, which is very surprising. And they're which is very surprising. And they're which is very surprising. And they're like, "Oh, yeah." And then we went to like, "Oh, yeah." And then we went to like, "Oh, yeah." And then we went to this thing called Turbopuffer, and they this thing called Turbopuffer, and they this thing called Turbopuffer, and they worked well. And I was like, what's worked well. And I was like, what's worked well. And I was like, what's turbuffer? And they're like, oh yeah, turbuffer? And they're like, oh yeah, turbuffer? And they're like, oh yeah, turbopuffer. I think I think they said turbopuffer. I think I think they said turbopuffer. I think I think they said like we were one of their first like we were one of their first like we were one of their first customers. And this never computed to customers. And this never computed to customers. And this never computed to me. Curser was already massive at that me. Curser was already massive at that me. Curser was already massive at that point. How did you meet the folks? And point. How did you meet the folks? And point. How did you meet the folks? And how did they become would were they the how did they become would were they the how did they become would were they the first customer? One of the first. first customer? One of the first. first customer? One of the first. >> They were the first customer. >> They were the first customer. >> They were the first customer. >> The first >> The first >> The first >> the first. >> the first. >> the first. >> No. >> No. >> No. >> Um they they they reached out um after I >> Um they they they reached out um after I >> Um they they they reached out um after I just launched on on Twitter. I was like, just launched on on Twitter. I was like, just launched on on Twitter. I was like, "Hey, I built this thing." And frankly "Hey, I built this thing." And frankly "Hey, I built this thing." And frankly it was like it was like it was like in I exact you it was like hey launch in I exact you it was like hey launch in I exact you it was like hey launch this thing and to me I was like I am so this thing and to me I was like I am so this thing and to me I was like I am so sick of working on this like I was like sick of working on this like I was like sick of working on this like I was like I've been working on this all summer I I've been working on this all summer I I've been working on this all summer I don't know if anyone cares I only want don't know if anyone cares I only want don't know if anyone cares I only want to work on this if anyone cares. Let's to work on this if anyone cares. Let's to work on this if anyone cares. Let's put it on Twitter again single Tox put it on Twitter again single Tox put it on Twitter again single Tox instance on a 8 core node somewhere in instance on a 8 core node somewhere in instance on a 8 core node somewhere in GCP. I was like if someone goes to prod GCP. I was like if someone goes to prod GCP. I was like if someone goes to prod I'll I'll set it up properly on multiple I'll I'll set it up properly on multiple I'll I'll set it up properly on multiple and like I'll just block on that but let and like I'll just block on that but let and like I'll just block on that but let let's see if anyone cares. It was like

let's see if anyone cares. It was like let's see if anyone cares. It was like the MVP of MVP. Anyone who's actually the MVP of MVP. Anyone who's actually the MVP of MVP. Anyone who's actually worked in the internal on databases worked in the internal on databases worked in the internal on databases would never have had like would have had would never have had like would have had would never have had like would have had too much pride to ship anything like too much pride to ship anything like too much pride to ship anything like that. that. that. Um, and I've just, you know, I've worked Um, and I've just, you know, I've worked Um, and I've just, you know, I've worked on I was just releasing it like a SAS on I was just releasing it like a SAS on I was just releasing it like a SAS project. Why can't you work on a project. Why can't you work on a project. Why can't you work on a database like it's SAS? I don't, you database like it's SAS? I don't, you database like it's SAS? I don't, you know, it's like if anyone uses it, we'll know, it's like if anyone uses it, we'll know, it's like if anyone uses it, we'll do it properly. I know how to run do it properly. I know how to run do it properly. I know how to run software with a lot of nines. Um, but it software with a lot of nines. Um, but it software with a lot of nines. Um, but it was not a proper LSN like it was very was not a proper LSN like it was very was not a proper LSN like it was very very it was the simplest version of what very it was the simplest version of what very it was the simplest version of what it could be. And then I released on it could be. And then I released on it could be. And then I released on Twitter. I was like, "Yeah, you could do Twitter. I was like, "Yeah, you could do Twitter. I was like, "Yeah, you could do a million vectors for a dollar." And a million vectors for a dollar." And a million vectors for a dollar." And before that, I think the the cheapest before that, I think the the cheapest before that, I think the the cheapest was maybe $100 per million for something was maybe $100 per million for something was maybe $100 per million for something that actually worked. that actually worked. that actually worked. >> Yeah. >> Yeah. >> Yeah. >> Um, and I knew it was reliable, right? I >> Um, and I knew it was reliable, right? I >> Um, and I knew it was reliable, right? I knew like I had these invariants like if knew like I had these invariants like if knew like I had these invariants like if you shut down all the VMs, like no data you shut down all the VMs, like no data you shut down all the VMs, like no data is lost, like all the rights are is lost, like all the rights are is lost, like all the rights are committed directly to object like it has committed directly to object like it has committed directly to object like it has all the same invariants it had today. Um all the same invariants it had today. Um all the same invariants it had today. Um and cursor reached out and knowing them and cursor reached out and knowing them and cursor reached out and knowing them now I'm sure at the time the cursor was now I'm sure at the time the cursor was now I'm sure at the time the cursor was maybe eight people and knowing the maybe eight people and knowing the maybe eight people and knowing the founders now I am sure that they had sat founders now I am sure that they had sat founders now I am sure that they had sat at the dinner table one day and we're at the dinner table one day and we're at the dinner table one day and we're like the unit economics of what we have like the unit economics of what we have like the unit economics of what we have right now where all the vectors are in right now where all the vectors are in right now where all the vectors are in DRAM are not working why hasn't anyone DRAM are not working why hasn't anyone DRAM are not working why hasn't anyone built it where we can put it in S3 and built it where we can put it in S3 and built it where we can put it in S3 and the actual code bases that are actively the actual code bases that are actively the actual code bases that are actively being used we can put in memory and being used we can put in memory and being used we can put in memory and everything else just sit in opic stores everything else just sit in opic stores everything else just sit in opic stores and then we just hotload it in and out and then we just hotload it in and out and then we just hotload it in and out of the cache of the cache of the cache >> makes so much sense right you open the >> makes so much sense right you open the >> makes so much sense right you open the codebase few seconds and it's in RAM and codebase few seconds and it's in RAM and codebase few seconds and it's in RAM and then the queries are as fast as anything then the queries are as fast as anything then the queries are as fast as anything else. It made so much sense. So I mean else. It made so much sense. So I mean else. It made so much sense. So I mean at the time they were if you look at at the time they were if you look at at the time they were if you look at some of Aman one of the co-founders some of Aman one of the co-founders some of Aman one of the co-founders early tweets he talks about uh using S3 early tweets he talks about uh using S3 early tweets he talks about uh using S3 for KV caching and things like that for KV caching and things like that for KV caching and things like that which barely anyone is still doing even

which barely anyone is still doing even which barely anyone is still doing even though though though >> um the economics >> um the economics >> um the economics >> yeah price wise yeah it's and it's it's >> yeah price wise yeah it's and it's it's >> yeah price wise yeah it's and it's it's very it's very uncommon and I think it very it's very uncommon and I think it very it's very uncommon and I think it will happen right but they were ahead of will happen right but they were ahead of will happen right but they were ahead of their time their time their time >> and they I think they were I don't know >> and they I think they were I don't know >> and they I think they were I don't know if they were thinking of building it if they were thinking of building it if they were thinking of building it themselves I think that's quite likely themselves I think that's quite likely themselves I think that's quite likely um and they found Turboper and it just um and they found Turboper and it just um and they found Turboper and it just perfectly pattern matched into that perfectly pattern matched into that perfectly pattern matched into that again I don't know if this dinner again I don't know if this dinner again I don't know if this dinner conversation happened or if this was conversation happened or if this was conversation happened or if this was just inside Harvey's head. Um, but it just inside Harvey's head. Um, but it just inside Harvey's head. Um, but it pattern matched something and so we pattern matched something and so we pattern matched something and so we exchanged a bunch of emails and then exchanged a bunch of emails and then exchanged a bunch of emails and then something compelled. I didn't know something compelled. I didn't know something compelled. I didn't know anything about B2B sales. anything about B2B sales. anything about B2B sales. >> Now I love B2B sales. Um, I didn't know >> Now I love B2B sales. Um, I didn't know >> Now I love B2B sales. Um, I didn't know anything. I was just like I just want to anything. I was just like I just want to anything. I was just like I just want to help them cuz they they were they had help them cuz they they were they had help them cuz they they were they had some unit economics that didn't line up. some unit economics that didn't line up. some unit economics that didn't line up. So I just went to San Francisco, right? So I just went to San Francisco, right? So I just went to San Francisco, right? I live in Canada. I went to San I live in Canada. I went to San I live in Canada. I went to San Francisco and I showed up at the office Francisco and I showed up at the office Francisco and I showed up at the office and when I showed up at the office they and when I showed up at the office they and when I showed up at the office they um they were having some Postgress um they were having some Postgress um they were having some Postgress problem that they were discussing. Yeah, problem that they were discussing. Yeah, problem that they were discussing. Yeah, the AWS aurora problems. Yes. the AWS aurora problems. Yes. the AWS aurora problems. Yes. >> Yeah. Early on. And I was like, "Oh, do >> Yeah. Early on. And I was like, "Oh, do >> Yeah. Early on. And I was like, "Oh, do you guys have PG analyze?" And they you guys have PG analyze?" And they you guys have PG analyze?" And they said, "Oh, no, we don't." I let's let's said, "Oh, no, we don't." I let's let's said, "Oh, no, we don't." I let's let's get that going, right? Let's look at it.

get that going, right? Let's look at it. get that going, right? Let's look at it. And it was the same thing as it always And it was the same thing as it always And it was the same thing as it always is with Postgress, which is autovacuum is with Postgress, which is autovacuum is with Postgress, which is autovacuum hadn't run enough and so they had all of hadn't run enough and so they had all of hadn't run enough and so they had all of these like going to heat when they these like going to heat when they these like going to heat when they should be doing index scans and blah should be doing index scans and blah should be doing index scans and blah blah blah. So, we were talking about all blah blah. So, we were talking about all blah blah. So, we were talking about all of that. And so, it's just helping them, of that. And so, it's just helping them, of that. And so, it's just helping them, right? It was like my, you know, my right? It was like my, you know, my right? It was like my, you know, my database genes just like kicked in. And database genes just like kicked in. And database genes just like kicked in. And I think this built enough trust with I think this built enough trust with I think this built enough trust with them that okay, well maybe if he knows them that okay, well maybe if he knows them that okay, well maybe if he knows how to help us with the database, maybe how to help us with the database, maybe how to help us with the database, maybe he also would know how to build one. And he also would know how to build one. And he also would know how to build one. And um at this time I'd also approached who um at this time I'd also approached who um at this time I'd also approached who I thought was the best engineer who ever I thought was the best engineer who ever I thought was the best engineer who ever worked at Shopify, my co-founder worked at Shopify, my co-founder worked at Shopify, my co-founder Justine. um and she'd come on and the Justine. um and she'd come on and the Justine. um and she'd come on and the first thing that she did was um remove first thing that she did was um remove first thing that she did was um remove the reverse proxy enginex cache with a the reverse proxy enginex cache with a the reverse proxy enginex cache with a file-based cache just a direct cache file-based cache just a direct cache file-based cache just a direct cache which again great like the S3 thing which again great like the S3 thing which again great like the S3 thing worked um and so she was online she was worked um and so she was online she was worked um and so she was online she was starting to work on it and um and cursor starting to work on it and um and cursor starting to work on it and um and cursor cursor cursor then that night was like cursor cursor then that night was like cursor cursor then that night was like okay well we're going to migrate and so okay well we're going to migrate and so okay well we're going to migrate and so they migrated everything over the course they migrated everything over the course they migrated everything over the course of like a week or two after that um but of like a week or two after that um but of like a week or two after that um but cursor was a small company back then cursor was a small company back then cursor was a small company back then right yeah and they were they just in right yeah and they were they just in right yeah and they were they just in the beginning of their massive rapid the beginning of their massive rapid the beginning of their massive rapid growth. Exactly. And I I told them that growth. Exactly. And I I told them that growth. Exactly. And I I told them that I was going to reduce their bill by 95%.

I was going to reduce their bill by 95%. I was going to reduce their bill by 95%. And I did like we did. Justine and I And I did like we did. Justine and I And I did like we did. Justine and I did. We like they came on and their last did. We like they came on and their last did. We like they came on and their last bill with their previous vendor and the bill with their previous vendor and the bill with their previous vendor and the first bill with us, it was 95% lower. first bill with us, it was 95% lower. first bill with us, it was 95% lower. Yeah. And you're you're nice for not Yeah. And you're you're nice for not Yeah. And you're you're nice for not saying vendors, but I I can say talks to saying vendors, but I I can say talks to saying vendors, but I I can say talks to them and and it's in the deep dive about them and and it's in the deep dive about them and and it's in the deep dive about cursor. It was it was a it was Aurora cursor. It was it was a it was Aurora cursor. It was it was a it was Aurora specifically. Uh so specifically. Uh so specifically. Uh so >> this was this was not this was not >> this was this was not this was not >> this was this was not this was not Postgress. No, this was a Postgress. No, this was a Postgress. No, this was a >> it was a different one, but it's >> it was a different one, but it's >> it was a different one, but it's probably still in the write up. We don't probably still in the write up. We don't probably still in the write up. We don't we don't need to name names, but uh we don't need to name names, but uh we don't need to name names, but uh yeah, but they were the reason they went yeah, but they were the reason they went yeah, but they were the reason they went there is reliability was their main main there is reliability was their main main there is reliability was their main main pain point. I'm sure the unit economics pain point. I'm sure the unit economics pain point. I'm sure the unit economics would have been there, but yeah, this would have been there, but yeah, this would have been there, but yeah, this was and then what Swallow told me is he was and then what Swallow told me is he was and then what Swallow told me is he said like look like there's a few things said like look like there's a few things said like look like there's a few things that we did never ever do and he said that we did never ever do and he said that we did never ever do and he said one of them you should never ever bet one of them you should never ever bet one of them you should never ever bet your business on a tiny startup where your business on a tiny startup where your business on a tiny startup where you are their only or biggest customer you are their only or biggest customer you are their only or biggest customer except for Turbopuffer and he said I except for Turbopuffer and he said I except for Turbopuffer and he said I love love those guys. So I guess it just love love those guys. So I guess it just love love those guys. So I guess it just comes to show that even in your case comes to show that even in your case comes to show that even in your case like to me what this story is shows is like to me what this story is shows is like to me what this story is shows is is you can do things when you build is you can do things when you build is you can do things when you build highquality things and you're pushing highquality things and you're pushing highquality things and you're pushing for things good things can happen and on for things good things can happen and on for things good things can happen and on the other side of cursor when you're a the other side of cursor when you're a the other side of cursor when you're a startup it's okay to take sometimes startup it's okay to take sometimes startup it's okay to take sometimes irrational risks when you have irrational risks when you have irrational risks when you have conviction and it sounds to me that you conviction and it sounds to me that you conviction and it sounds to me that you gave them conviction by showing up in gave them conviction by showing up in gave them conviction by showing up in person by helping them by showing that person by helping them by showing that person by helping them by showing that you know you know your stuff like you you know you know your stuff like you you know you know your stuff like you suddenly brought in your your 10ish or suddenly brought in your your 10ish or suddenly brought in your your 10ish or eight years of Shopify experience and eight years of Shopify experience and eight years of Shopify experience and your curiosity and They probably took a your curiosity and They probably took a your curiosity and They probably took a risk because of that, not because you risk because of that, not because you risk because of that, not because you were some, you know, random vendor. They were some, you know, random vendor. They were some, you know, random vendor. They probably never done that. So, fast probably never done that. So, fast probably never done that. So, fast forward today, uh, Turbopuffer is now a forward today, uh, Turbopuffer is now a forward today, uh, Turbopuffer is now a lot bigger. You're you're working on lot bigger. You're you're working on lot bigger. You're you're working on some some some cool things, but

some some some cool things, but some some some cool things, but you have this very interesting business you have this very interesting business you have this very interesting business where for you CPUs are important, right? where for you CPUs are important, right? where for you CPUs are important, right? You run on mostly CPUs. And you told me You run on mostly CPUs. And you told me You run on mostly CPUs. And you told me a story over dinner yesterday that a story over dinner yesterday that a story over dinner yesterday that uh you met Jensen uh and Jensen he uh you met Jensen uh and Jensen he uh you met Jensen uh and Jensen he really wanted to sell you on GPUs. Can really wanted to sell you on GPUs. Can really wanted to sell you on GPUs. Can you tell me how that meeting went? you tell me how that meeting went? you tell me how that meeting went? >> Um yeah, Jensen Hong, right? >> Um yeah, Jensen Hong, right? >> Um yeah, Jensen Hong, right? >> Yeah. I just I never met uh I'd never >> Yeah. I just I never met uh I'd never >> Yeah. I just I never met uh I'd never met uh Jensen before. We were we were at met uh Jensen before. We were we were at met uh Jensen before. We were we were at an event at uh at at Nvidia and we were an event at uh at at Nvidia and we were an event at uh at at Nvidia and we were just doing um presentations in a big HQ. just doing um presentations in a big HQ. just doing um presentations in a big HQ. Super impressive. Super impressive. Super impressive. >> Yeah, exactly. they've invited a couple >> Yeah, exactly. they've invited a couple >> Yeah, exactly. they've invited a couple of companies to go and and um and and of companies to go and and um and and of companies to go and and um and and and talk about um uh talk about our and talk about um uh talk about our and talk about um uh talk about our businesses and how we can partner with businesses and how we can partner with businesses and how we can partner with Nvidia and so on. And Nvidia and so on. And Nvidia and so on. And I I don't I I don't know. I was like I I I don't I I don't know. I was like I I I don't I I don't know. I was like I think I was in a goofy mood that day. think I was in a goofy mood that day. think I was in a goofy mood that day. And so I went up on on stage and I said, And so I went up on on stage and I said, And so I went up on on stage and I said, "Um, hey, I'm Simon from from "Um, hey, I'm Simon from from "Um, hey, I'm Simon from from Turbopuffer." And uh and yeah, if you're Turbopuffer." And uh and yeah, if you're Turbopuffer." And uh and yeah, if you're wondering about the name, it's like if wondering about the name, it's like if wondering about the name, it's like if everything goes south, we can always everything goes south, we can always everything goes south, we can always pivot into vapes.

pivot into vapes. pivot into vapes. >> [laughter] >> [laughter] >> [laughter] >> I was kind of nervous and this is what I >> I was kind of nervous and this is what I >> I was kind of nervous and this is what I this is what I said and then and then he this is what I said and then and then he this is what I said and then and then he said back to said back to said back to >> wait who was in the room? Was it Jensen? >> wait who was in the room? Was it Jensen? >> wait who was in the room? Was it Jensen? Was it a direct report? Was it a direct report? Was it a direct report? >> It was it was Jensen and then I don't he >> It was it was Jensen and then I don't he >> It was it was Jensen and then I don't he has like I don't know if it's just 50 has like I don't know if it's just 50 has like I don't know if it's just 50 direct reports or it was like you know direct reports or it was like you know direct reports or it was like you know it was there was it was Jensen and then it was there was it was Jensen and then it was there was it was Jensen and then a bunch of the um like Nvidia Nvidia a bunch of the um like Nvidia Nvidia a bunch of the um like Nvidia Nvidia leadership, right? Um cuz you go there leadership, right? Um cuz you go there leadership, right? Um cuz you go there and then you talk about that you find and then you talk about that you find and then you talk about that you find opportunities to partner and work opportunities to partner and work opportunities to partner and work together, right? And so I said, "Yeah, together, right? And so I said, "Yeah, together, right? And so I said, "Yeah, you know, so plan B could be that we you know, so plan B could be that we you know, so plan B could be that we could pivot into vapes." And then he could pivot into vapes." And then he could pivot into vapes." And then he said, I was already nervous. He said, said, I was already nervous. He said, said, I was already nervous. He said, "Judging by your slide, maybe you "Judging by your slide, maybe you "Judging by your slide, maybe you should." should." should." [laughter] [laughter] [laughter] No, he did not. No, he did not. No, he did not. [laughter] [laughter] [laughter] And and And and And and [gasps] [gasps] [gasps] I didn't know what to say back to that. I didn't know what to say back to that. I didn't know what to say back to that. So I said, "Well, Jensen, do you vape?" [snorts] [snorts] [laughter] [laughter] [laughter] He didn't he didn't answer the question.

He didn't he didn't answer the question. He didn't he didn't answer the question. [laughter] And then someone um someone [laughter] And then someone um someone [laughter] And then someone um someone on the um someone on the on the team um on the um someone on the on the team um on the um someone on the on the team um wrote to the whole company, Turop Puffer wrote to the whole company, Turop Puffer wrote to the whole company, Turop Puffer Company, Simon just asked Jensen if he Company, Simon just asked Jensen if he Company, Simon just asked Jensen if he vapes. vapes. vapes. Um, and then you know this is this is a Um, and then you know this is this is a Um, and then you know this is this is a great start, right? And um, and then the great start, right? And um, and then the great start, right? And um, and then the team had team team had sort of talked to team had team team had sort of talked to team had team team had sort of talked to me beforehand. I was like, Simon, we got me beforehand. I was like, Simon, we got me beforehand. I was like, Simon, we got to make sure we don't say the C-word. We to make sure we don't say the C-word. We to make sure we don't say the C-word. We can't say CPUs. And so I just couldn't stop talking And so I just couldn't stop talking about CPUs. I was like, AVX 512 is so about CPUs. I was like, AVX 512 is so about CPUs. I was like, AVX 512 is so sick. Like we love SIMD and um, like we sick. Like we love SIMD and um, like we sick. Like we love SIMD and um, like we we we like there's so many CPUs. They're we we like there's so many CPUs. They're we we like there's so many CPUs. They're so easy to get. like um it's just a riot so easy to get. like um it's just a riot so easy to get. like um it's just a riot in CPU land. Like you know I I don't I in CPU land. Like you know I I don't I in CPU land. Like you know I I don't I think I stopped short of saying I'm so think I stopped short of saying I'm so think I stopped short of saying I'm so glad I don't need GPUs. [laughter] But glad I don't need GPUs. [laughter] But glad I don't need GPUs. [laughter] But but it was just it I just couldn't stop but it was just it I just couldn't stop but it was just it I just couldn't stop talking about CPUs. talking about CPUs. talking about CPUs. Yeah. And so you know Jensen took an Yeah. And so you know Jensen took an Yeah. And so you know Jensen took an interest in that. Yeah. So who who knows interest in that. Yeah. So who who knows interest in that. Yeah. So who who knows like I'm sure you made you made a like I'm sure you made you made a like I'm sure you made you made a memorable impression. Maybe he made it memorable impression. Maybe he made it memorable impression. Maybe he made it his mission now to like at some point his mission now to like at some point his mission now to like at some point get you guys onto GPUs. But speaking of get you guys onto GPUs. But speaking of get you guys onto GPUs. But speaking of CPUs, can you tell me a bit what you're CPUs, can you tell me a bit what you're CPUs, can you tell me a bit what you're seeing inside of the hypers scale, the seeing inside of the hypers scale, the seeing inside of the hypers scale, the cloud providers? You're now in AWS, cloud providers? You're now in AWS, cloud providers? You're now in AWS, you're you're in GCP, you're on Azure.

you're you're in GCP, you're on Azure. you're you're in GCP, you're on Azure. What I would think naively is there's a What I would think naively is there's a What I would think naively is there's a GPU shortage and when I talk with GPU shortage and when I talk with GPU shortage and when I talk with inference companies, they are and and inference companies, they are and and inference companies, they are and and and AI labs, they're just getting and AI labs, they're just getting and AI labs, they're just getting whatever they can do. I would think whatever they can do. I would think whatever they can do. I would think getting CPUs is should be easy. Is it? getting CPUs is should be easy. Is it? getting CPUs is should be easy. Is it? >> No, >> No, >> No, >> it's not anymore. Why? What's happening? >> it's not anymore. Why? What's happening? >> it's not anymore. Why? What's happening? Can you tell us about dynamics on on on Can you tell us about dynamics on on on Can you tell us about dynamics on on on the why and what you've learned? Yeah. the why and what you've learned? Yeah. the why and what you've learned? Yeah. So, I think that So, I think that So, I think that GPUs will probably continue to be GPUs will probably continue to be GPUs will probably continue to be scarce. Like, I don't know, maybe scarce. Like, I don't know, maybe scarce. Like, I don't know, maybe there's going to be some surplus. I I there's going to be some surplus. I I there's going to be some surplus. I I refuse to speculate too much about the refuse to speculate too much about the refuse to speculate too much about the macro, but I think as as RL is becoming macro, but I think as as RL is becoming macro, but I think as as RL is becoming a very very large amount of the a very very large amount of the a very very large amount of the workloads that needs a lot of CPUs. So, workloads that needs a lot of CPUs. So, workloads that needs a lot of CPUs. So, the labs are sucking up a lot of CPUs the labs are sucking up a lot of CPUs the labs are sucking up a lot of CPUs because you need CPUs to be like, okay, because you need CPUs to be like, okay, because you need CPUs to be like, okay, we need to like teach this model how to we need to like teach this model how to we need to like teach this model how to how to search. We need to teach it how how to search. We need to teach it how how to search. We need to teach it how to use GP. We need to teach it how to to use GP. We need to teach it how to to use GP. We need to teach it how to boot up bash. We need it needs to run boot up bash. We need it needs to run boot up bash. We need it needs to run real things and learn from that takes a real things and learn from that takes a real things and learn from that takes a lot of CPU. lot of CPU. lot of CPU. >> Um and so I think as we RL is consuming >> Um and so I think as we RL is consuming >> Um and so I think as we RL is consuming a lot of CPU and then also just all of a lot of CPU and then also just all of a lot of CPU and then also just all of the agents are running on CPUs, right? the agents are running on CPUs, right? the agents are running on CPUs, right? They need to do all kinds of very They need to do all kinds of very They need to do all kinds of very general purpose things on a CPU and so general purpose things on a CPU and so general purpose things on a CPU and so as as as the demand curve is sort of as as as the demand curve is sort of as as as the demand curve is sort of shifting to the right and it's becoming shifting to the right and it's becoming shifting to the right and it's becoming more and more applied and that feeds more and more applied and that feeds more and more applied and that feeds back into RL by the way, right? because back into RL by the way, right? because back into RL by the way, right? because as things become more applied like oh as things become more applied like oh as things become more applied like oh the models are not that good at CAD or the models are not that good at CAD or the models are not that good at CAD or ship building I don't know and then you ship building I don't know and then you ship building I don't know and then you know you have to spin up even more RL know you have to spin up even more RL know you have to spin up even more RL environments to do that so I think environments to do that so I think environments to do that so I think that's what we're seeing and so we're on that's what we're seeing and so we're on that's what we're seeing and so we're on the other end of that needing these CPUs the other end of that needing these CPUs the other end of that needing these CPUs we need a lot of NVME SSDs as well um we need a lot of NVME SSDs as well um we need a lot of NVME SSDs as well um and a lot of this right now is tied up and a lot of this right now is tied up and a lot of this right now is tied up in DRAM right of of where like

in DRAM right of of where like in DRAM right of of where like >> you need a lot of that also for the GPU >> you need a lot of that also for the GPU >> you need a lot of that also for the GPU servers um but I would assume that it servers um but I would assume that it servers um but I would assume that it gets a lot worse before it gets a lot gets a lot worse before it gets a lot gets a lot worse before it gets a lot better on the on the CPU side um and I better on the on the CPU side um and I better on the on the CPU side um and I think even the big companies are think even the big companies are think even the big companies are fighting amongst each other, right, to fighting amongst each other, right, to fighting amongst each other, right, to get the allocations and even we, you get the allocations and even we, you get the allocations and even we, you know, we're selling to companies that we know, we're selling to companies that we know, we're selling to companies that we also fight for CPU with and against, also fight for CPU with and against, also fight for CPU with and against, right? It's uh it's it's it's really right? It's uh it's it's it's really right? It's uh it's it's it's really difficult and so you write things to try difficult and so you write things to try difficult and so you write things to try to make sure you get these CPUs as f to make sure you get these CPUs as f to make sure you get these CPUs as f fast as possible. fast as possible. fast as possible. >> Yeah. And yesterday I was at a dinner >> Yeah. And yesterday I was at a dinner >> Yeah. And yesterday I was at a dinner that you hosted with your team where you that you hosted with your team where you that you hosted with your team where you actually have a bunch of Turbo actually have a bunch of Turbo actually have a bunch of Turbo customers. A bunch of them are AI AI customers. A bunch of them are AI AI customers. A bunch of them are AI AI labs or or AI startups but a lot of them labs or or AI startups but a lot of them labs or or AI startups but a lot of them one of them uh reflection had have hu one of them uh reflection had have hu one of them uh reflection had have hu massive amount of footprint and they massive amount of footprint and they massive amount of footprint and they were telling me that they're in a were telling me that they're in a were telling me that they're in a situation where they cannot buy more situation where they cannot buy more situation where they cannot buy more like they when it comes to GPUs or CPUs like they when it comes to GPUs or CPUs like they when it comes to GPUs or CPUs they max out they have the longest they max out they have the longest they max out they have the longest contracts that possible and I didn't contracts that possible and I didn't contracts that possible and I didn't realize how competitive it is in the realize how competitive it is in the realize how competitive it is in the cloud when you you go beyond a small cloud when you you go beyond a small cloud when you you go beyond a small fish to like a medium size or even a a fish to like a medium size or even a a fish to like a medium size or even a a large fish that now like It's it's large fish that now like It's it's large fish that now like It's it's interesting. So So now you have this and interesting. So So now you have this and interesting. So So now you have this and even you're having this this uh kind of even you're having this this uh kind of even you're having this this uh kind of fight behind the scenes that is maybe fight behind the scenes that is maybe fight behind the scenes that is maybe not as visible.

not as visible. not as visible. >> Exactly. And I mean you you work with >> Exactly. And I mean you you work with >> Exactly. And I mean you you work with the clouds right you work with them to the clouds right you work with them to the clouds right you work with them to talk about which regions have um have talk about which regions have um have talk about which regions have um have CPU which regions are getting it comes CPU which regions are getting it comes CPU which regions are getting it comes down to power right of like okay well down to power right of like okay well down to power right of like okay well where is the power which is generally where is the power which is generally where is the power which is generally where they're going to ship the new where they're going to ship the new where they're going to ship the new CPUs. Um and so we have to work with CPUs. Um and so we have to work with CPUs. Um and so we have to work with some of our biggest customers on that. some of our biggest customers on that. some of our biggest customers on that. So these are real constraints right that So these are real constraints right that So these are real constraints right that are that are making our way to us. We're are that are making our way to us. We're are that are making our way to us. We're just very fortunate that it's very easy just very fortunate that it's very easy just very fortunate that it's very easy for us to run lots of turbuffer clusters for us to run lots of turbuffer clusters for us to run lots of turbuffer clusters because all we need are like a few CPUs because all we need are like a few CPUs because all we need are like a few CPUs and NVME SSDs and then S3 and then we're and NVME SSDs and then S3 and then we're and NVME SSDs and then S3 and then we're in a good place. But there's lots of in a good place. But there's lots of in a good place. But there's lots of changes that we can make even to the changes that we can make even to the changes that we can make even to the architecture um to try to protect from architecture um to try to protect from architecture um to try to protect from from a lot of this. Now I'd rather spend from a lot of this. Now I'd rather spend from a lot of this. Now I'd rather spend that engineering effort on other things that engineering effort on other things that engineering effort on other things but we are very very good at using a lot but we are very very good at using a lot but we are very very good at using a lot of very different SKs, right? So we of very different SKs, right? So we of very different SKs, right? So we don't need everything to be a particular don't need everything to be a particular don't need everything to be a particular CPU or instance type. We can run with CPU or instance type. We can run with CPU or instance type. We can run with many even many different types of many even many different types of many even many different types of machine types. Um machine types. Um machine types. Um >> Q meaning that's the it's a fancy name >> Q meaning that's the it's a fancy name >> Q meaning that's the it's a fancy name for like the different machine types. for like the different machine types. for like the different machine types. >> Yes. Exactly. Right. Like you know C4D >> Yes. Exactly. Right. Like you know C4D >> Yes. Exactly. Right. Like you know C4D or I AG or whatever they're called.

or I AG or whatever they're called. or I AG or whatever they're called. >> What's your favorite one? >> What's your favorite one? >> What's your favorite one? >> Um we really like right now the um C4s >> Um we really like right now the um C4s >> Um we really like right now the um C4s on uh GCP. on uh GCP. on uh GCP. >> GCP. >> GCP. >> GCP. >> Um the Z4Ds are also performing really >> Um the Z4Ds are also performing really >> Um the Z4Ds are also performing really well um now that we've done done a bunch well um now that we've done done a bunch well um now that we've done done a bunch of of um of optimizations to them. Um of of um of optimizations to them. Um of of um of optimizations to them. Um those are really really great machine those are really really great machine those are really really great machine types. Uh we really like those. Um and types. Uh we really like those. Um and types. Uh we really like those. Um and then the ARM C4As as well um on on GCP. then the ARM C4As as well um on on GCP. then the ARM C4As as well um on on GCP. Um we like those. But I think that in Um we like those. But I think that in Um we like those. But I think that in general like when you're yeah when general like when you're yeah when general like when you're yeah when you're small it's very easy to suck up a you're small it's very easy to suck up a you're small it's very easy to suck up a bunch of but at Shopify I was also part bunch of but at Shopify I was also part bunch of but at Shopify I was also part of you know deciding of you know deciding of you know deciding ahead of BFCM right a few months out you ahead of BFCM right a few months out you ahead of BFCM right a few months out you have to tell the cloud providers how have to tell the cloud providers how have to tell the cloud providers how much you're intending to use. do commits much you're intending to use. do commits much you're intending to use. do commits on all of that, right? The the clouds on all of that, right? The the clouds on all of that, right? The the clouds are not infinite as they seem when are not infinite as they seem when are not infinite as they seem when you're small. you're small. you're small. >> And one way, of course, to like get like >> And one way, of course, to like get like >> And one way, of course, to like get like infrastructure and and also just like infrastructure and and also just like infrastructure and and also just like credibility is venture capital. If you credibility is venture capital. If you credibility is venture capital. If you raise $100 million, a billion dollars, raise $100 million, a billion dollars, raise $100 million, a billion dollars, some of your customers just raised $2 some of your customers just raised $2 some of your customers just raised $2 billion. Actually, I talked with them billion. Actually, I talked with them billion. Actually, I talked with them yesterday. You know, it gives you yesterday. You know, it gives you yesterday. You know, it gives you credibility, gives you cash, you can pay credibility, gives you cash, you can pay credibility, gives you cash, you can pay for this thing. Your specific for this thing. Your specific for this thing. Your specific Turbopufferers relationship to venture Turbopufferers relationship to venture Turbopufferers relationship to venture capital seems very interesting. I never capital seems very interesting. I never capital seems very interesting. I never heard you announce a raise until m maybe heard you announce a raise until m maybe heard you announce a raise until m maybe just very recently. Can you tell me how just very recently. Can you tell me how just very recently. Can you tell me how you and and you told me that when you you and and you told me that when you you and and you told me that when you started this thing, you didn't think too started this thing, you didn't think too started this thing, you didn't think too much outside of just building some cool much outside of just building some cool much outside of just building some cool stuff. How did you think about venture stuff. How did you think about venture stuff. How did you think about venture capital and how do you think about capital and how do you think about capital and how do you think about raising because again I feel you have a raising because again I feel you have a raising because again I feel you have a very fresh and different perspective very fresh and different perspective very fresh and different perspective than what which is typical inside of than what which is typical inside of than what which is typical inside of Silicon Valley. Yeah. So I think to to

Silicon Valley. Yeah. So I think to to Silicon Valley. Yeah. So I think to to understand my how I think about capital understand my how I think about capital understand my how I think about capital you have to go back to the the beginning you have to go back to the the beginning you have to go back to the the beginning of Turbopuffer, right? where I promised of Turbopuffer, right? where I promised of Turbopuffer, right? where I promised cursor that Justine and I could get cursor that Justine and I could get cursor that Justine and I could get their bill to 4K a month. And this was their bill to 4K a month. And this was their bill to 4K a month. And this was based on some very rough napkin math on, based on some very rough napkin math on, based on some very rough napkin math on, okay, if if Turbopuffer was a better okay, if if Turbopuffer was a better okay, if if Turbopuffer was a better implementation than it currently is, implementation than it currently is, implementation than it currently is, then it should cost this much. And then it should cost this much. And then it should cost this much. And that's the pricing we ship with and that's the pricing we ship with and that's the pricing we ship with and that's what we guaranteed um guaranteed that's what we guaranteed um guaranteed that's what we guaranteed um guaranteed cursor. Um but the software was not that cursor. Um but the software was not that cursor. Um but the software was not that good. Like it was very reliable, but it good. Like it was very reliable, but it good. Like it was very reliable, but it was very simple, right? And that's like was very simple, right? And that's like was very simple, right? And that's like a core engineering principle of me is a core engineering principle of me is a core engineering principle of me is simplicity above everything. Um you and simplicity above everything. Um you and simplicity above everything. Um you and I have talked before about how software I have talked before about how software I have talked before about how software that ages well and some of the that ages well and some of the that ages well and some of the advantages of seeing be having long advantages of seeing be having long advantages of seeing be having long tenures inside of companies. You had a tenures inside of companies. You had a tenures inside of companies. You had a long tenure at Uber. I had a long tenure long tenure at Uber. I had a long tenure long tenure at Uber. I had a long tenure at Shopify. So you see simplicity just at Shopify. So you see simplicity just at Shopify. So you see simplicity just almost always wins. Um and at the time I almost always wins. Um and at the time I almost always wins. Um and at the time I was not convinced whether this was a was not convinced whether this was a was not convinced whether this was a venture scale opportunity because I venture scale opportunity because I venture scale opportunity because I understood that if you take venture understood that if you take venture understood that if you take venture capital no matter how many smiles there capital no matter how many smiles there capital no matter how many smiles there are in the room everyone's sort of are in the room everyone's sort of are in the room everyone's sort of expecting that you have to earn a big expecting that you have to earn a big expecting that you have to earn a big return on that on some timeline that return on that on some timeline that return on that on some timeline that makes sense to everyone involved and makes sense to everyone involved and makes sense to everyone involved and everyone involved are you know pension everyone involved are you know pension everyone involved are you know pension funds in Canada like that it's like it funds in Canada like that it's like it funds in Canada like that it's like it there's like a whole stack right of of there's like a whole stack right of of there's like a whole stack right of of of people that that need to so at the of people that that need to so at the of people that that need to so at the time I was like I don't you I don't know time I was like I don't you I don't know time I was like I don't you I don't know if this could be a billion dollar if this could be a billion dollar if this could be a billion dollar company. I didn't know that in the very company. I didn't know that in the very company. I didn't know that in the very very beginning. Um it wasn't completely very beginning. Um it wasn't completely very beginning. Um it wasn't completely clear to me. It felt like a very niche clear to me. It felt like a very niche clear to me. It felt like a very niche kind of product, right, to build this kind of product, right, to build this kind of product, right, to build this particular search engine. Um and that

particular search engine. Um and that particular search engine. Um and that was completely fine with me. So I you was completely fine with me. So I you was completely fine with me. So I you know it's it's it was fine. And know it's it's it was fine. And know it's it's it was fine. And so then I just I just looked at the so then I just I just looked at the so then I just I just looked at the cursor bill and I looked at my GCP bill cursor bill and I looked at my GCP bill cursor bill and I looked at my GCP bill which is what we started on. and you which is what we started on. and you which is what we started on. and you know as like a you know dumb Danish know as like a you know dumb Danish know as like a you know dumb Danish person who's just like okay like this person who's just like okay like this person who's just like okay like this number should just be lower than the number should just be lower than the number should just be lower than the other number. other number. other number. >> Yeah, >> Yeah, >> Yeah, >> that's sort of like you know and it's >> that's sort of like you know and it's >> that's sort of like you know and it's just I don't think I'd spend enough time just I don't think I'd spend enough time just I don't think I'd spend enough time in San Francisco cuz I think the money in San Francisco cuz I think the money in San Francisco cuz I think the money over here it works a little bit over here it works a little bit over here it works a little bit differently. Um that's just that's all I differently. Um that's just that's all I differently. Um that's just that's all I knew. You you were doing business 101 as knew. You you were doing business 101 as knew. You you were doing business 101 as as long as you're making a profit you're as long as you're making a profit you're as long as you're making a profit you're good, right? good, right? good, right? >> Yeah. >> Yeah. >> Yeah. It was like I'm I'm not kidding in this It was like I'm I'm not kidding in this It was like I'm I'm not kidding in this exaggeration that it was just like that exaggeration that it was just like that exaggeration that it was just like that just made sense to me that Justine and I just made sense to me that Justine and I just made sense to me that Justine and I were just going to go optimize this were just going to go optimize this were just going to go optimize this until these numbers were roughly equal. until these numbers were roughly equal. until these numbers were roughly equal. And maybe if if if we could get some And maybe if if if we could get some And maybe if if if we could get some other workloads, we could start paying other workloads, we could start paying other workloads, we could start paying ourselves. But that was like very much ourselves. But that was like very much ourselves. But that was like very much the philosophy at the time. Um because I the philosophy at the time. Um because I the philosophy at the time. Um because I didn't know if I could go raise a bunch didn't know if I could go raise a bunch didn't know if I could go raise a bunch of of of money. I didn't know anyone who of of of money. I didn't know anyone who of of of money. I didn't know anyone who had the money. I I didn't have any had the money. I I didn't have any had the money. I I didn't have any relationships. Um you were an absolute relationships. Um you were an absolute relationships. Um you were an absolute outsider to the outsider to the outsider to the >> I was I was an outsider. I was like an >> I was I was an outsider. I was like an >> I was I was an outsider. I was like an outsider squared, right? I grew up in outsider squared, right? I grew up in outsider squared, right? I grew up in Aus, Denmark and I um I then moved to Aus, Denmark and I um I then moved to Aus, Denmark and I um I then moved to Ottawa, Canada. So it's like I'm an Ottawa, Canada. So it's like I'm an Ottawa, Canada. So it's like I'm an outsider to Canada and in Canada I'm an outsider to Canada and in Canada I'm an outsider to Canada and in Canada I'm an outsider to San Francisco. So I was just outsider to San Francisco. So I was just outsider to San Francisco. So I was just thinking about this from first thinking about this from first thinking about this from first principles like oh you're a venture principles like oh you're a venture principles like oh you're a venture capital you need this return you need it capital you need this return you need it capital you need this return you need it on this timeline.

on this timeline. on this timeline. I don't know if I can deliver that yet. I don't know if I can deliver that yet. I don't know if I can deliver that yet. I would need more data to decide that I would need more data to decide that I would need more data to decide that because I want to like I kind of want to because I want to like I kind of want to because I want to like I kind of want to keep working on this and now I have to keep working on this and now I have to keep working on this and now I have to get to this point for it to not be a get to this point for it to not be a get to this point for it to not be a failure. Um in in in January then I uh failure. Um in in in January then I uh failure. Um in in in January then I uh there was a person that I was at II with there was a person that I was at II with there was a person that I was at II with in in uh in 2012 and 2013 and his name in in uh in 2012 and 2013 and his name in in uh in 2012 and 2013 and his name is Buen and he was on the North and is Buen and he was on the North and is Buen and he was on the North and Macedonian team um at II um and he was Macedonian team um at II um and he was Macedonian team um at II um and he was he's he was really good. He was so good he's he was really good. He was so good he's he was really good. He was so good that the North Macedonian team called that the North Macedonian team called that the North Macedonian team called him God. Um him God. Um him God. Um I don't know why but that was what he I don't know why but that was what he I don't know why but that was what he went by and he was yeah he was very good went by and he was yeah he was very good went by and he was yeah he was very good grew up and and I really wanted to work grew up and and I really wanted to work grew up and and I really wanted to work with Buen but I couldn't afford to work with Buen but I couldn't afford to work with Buen but I couldn't afford to work with Boyan [laughter] with Boyan [laughter] with Boyan [laughter] um and he was very much like this is um and he was very much like this is um and he was very much like this is what I can live off like you know I just what I can live off like you know I just what I can live off like you know I just like I want to build this date like that like I want to build this date like that like I want to build this date like that would be like this is what it can be but would be like this is what it can be but would be like this is what it can be but at this point Justine and I hadn't taken at this point Justine and I hadn't taken at this point Justine and I hadn't taken a salary for like 6 months and we'd a salary for like 6 months and we'd a salary for like 6 months and we'd already we'd already spent like tens of already we'd already spent like tens of already we'd already spent like tens of thousands of dollars on like on GCP thousands of dollars on like on GCP thousands of dollars on like on GCP bills and all of that and I was like I bills and all of that and I was like I bills and all of that and I was like I don't think we can I don't think we can don't think we can I don't think we can don't think we can I don't think we can we we can do it. And so I had met one we we can do it. And so I had met one we we can do it. And so I had met one one individual in in Silicon Valley uh one individual in in Silicon Valley uh one individual in in Silicon Valley uh his name is Locky and it just I ended up his name is Locky and it just I ended up his name is Locky and it just I ended up just calling him and saying hey I kind just calling him and saying hey I kind just calling him and saying hey I kind of want to learn a little bit faster of want to learn a little bit faster of want to learn a little bit faster here.

here. here. Can I can we raise like 700K? That's Can I can we raise like 700K? That's Can I can we raise like 700K? That's like what I wanted to raise. So it's like what I wanted to raise. So it's like what I wanted to raise. So it's just like I want to have like two just like I want to have like two just like I want to have like two engineers for the rest of the year. engineers for the rest of the year. engineers for the rest of the year. Justine and I still don't need to be Justine and I still don't need to be Justine and I still don't need to be paid and then a little bit of buffer paid and then a little bit of buffer paid and then a little bit of buffer room. It's like this is what I need and room. It's like this is what I need and room. It's like this is what I need and if this doesn't have PMF and is a big if this doesn't have PMF and is a big if this doesn't have PMF and is a big opportunity by the end of the year, I opportunity by the end of the year, I opportunity by the end of the year, I don't think we're going to bother and don't think we're going to bother and don't think we're going to bother and we'll just shut the whole thing down and we'll just shut the whole thing down and we'll just shut the whole thing down and we won't have it taking a dime. We'll we won't have it taking a dime. We'll we won't have it taking a dime. We'll return everything to you. Um return everything to you. Um return everything to you. Um I think there was the first time you I think there was the first time you I think there was the first time you heard anyone say it like that. Um and um heard anyone say it like that. Um and um heard anyone say it like that. Um and um I told some other VCs that at the time I told some other VCs that at the time I told some other VCs that at the time and that was terrifying to them. I think and that was terrifying to them. I think and that was terrifying to them. I think to someone on the West Coast this sounds to someone on the West Coast this sounds to someone on the West Coast this sounds like you have low ambition or something like you have low ambition or something like you have low ambition or something like that. M like that. M like that. M >> um and to me it was just like I I don't >> um and to me it was just like I I don't >> um and to me it was just like I I don't know it just came from a when I don't know it just came from a when I don't know it just came from a when I don't know how to play a game I just play with know how to play a game I just play with know how to play a game I just play with open cards like this is how I see it open cards like this is how I see it open cards like this is how I see it >> and so I we were it was very clear to us >> and so I we were it was very clear to us >> and so I we were it was very clear to us that we wanted to do this and but also that we wanted to do this and but also that we wanted to do this and but also it became clear to us that we didn't it became clear to us that we didn't it became clear to us that we didn't want to just like keep working on this want to just like keep working on this want to just like keep working on this unless it could become big and we were unless it could become big and we were unless it could become big and we were starting to develop conviction starting to develop conviction starting to develop conviction conviction that this actually become conviction that this actually become conviction that this actually become really really big and so we we we did really really big and so we we we did really really big and so we we we did that and hired Buen and then became that and hired Buen and then became that and hired Buen and then became profitable later that year um and then profitable later that year um and then profitable later that year um and then just continued to hire and then it's just continued to hire and then it's just continued to hire and then it's like to raise more money, you need sort like to raise more money, you need sort like to raise more money, you need sort of there's six reasons to raise capital.

of there's six reasons to raise capital. of there's six reasons to raise capital. The first reason to raise capital is to The first reason to raise capital is to The first reason to raise capital is to fund R&D. fund R&D. fund R&D. >> Mhm. >> Mhm. >> Mhm. >> That was the reason that we raised >> That was the reason that we raised >> That was the reason that we raised capital in January because we funded R&D capital in January because we funded R&D capital in January because we funded R&D with a lot of our own, you know, with a lot of our own, you know, with a lot of our own, you know, opportunity cost and not taking a salary opportunity cost and not taking a salary opportunity cost and not taking a salary and then paying the bills ourselves. Um, and then paying the bills ourselves. Um, and then paying the bills ourselves. Um, but we wanted to learn a little bit but we wanted to learn a little bit but we wanted to learn a little bit faster and so we hired Buen and Morgan faster and so we hired Buen and Morgan faster and so we hired Buen and Morgan as the first engineers. And then the as the first engineers. And then the as the first engineers. And then the second reason to raise capital is to second reason to raise capital is to second reason to raise capital is to fund growth. you've you you've built fund growth. you've you you've built fund growth. you've you you've built something and you want to tell the world something and you want to tell the world something and you want to tell the world about it and you want to spend more about it and you want to spend more about it and you want to spend more capital to do that. Um the third reason capital to do that. Um the third reason capital to do that. Um the third reason to to to raise capital is for the to to to raise capital is for the to to to raise capital is for the founders's ego. [snorts] founders's ego. [snorts] founders's ego. [snorts] >> Um it's a very popular appreciate the >> Um it's a very popular appreciate the >> Um it's a very popular appreciate the honesty. honesty. honesty. >> It's very popular, very very popular, >> It's very popular, very very popular, >> It's very popular, very very popular, right? big numbers, lots of press, like right? big numbers, lots of press, like right? big numbers, lots of press, like um and I think this is a very very um and I think this is a very very um and I think this is a very very dangerous reason to raise money. And I dangerous reason to raise money. And I dangerous reason to raise money. And I wish that it was more talked about wish that it was more talked about wish that it was more talked about because you're diluting all of your because you're diluting all of your because you're diluting all of your employees when you do it. You are um employees when you do it. You are um employees when you do it. You are um setting a certain price for future setting a certain price for future setting a certain price for future employees and their upside. It's employees and their upside. It's employees and their upside. It's it's it for some people it can become a it's it for some people it can become a it's it for some people it can become a status game and that's not what it's status game and that's not what it's status game and that's not what it's about. We're here to build a big about. We're here to build a big about. We're here to build a big business together and business together and business together and this is not a reason to raise money. Um, this is not a reason to raise money. Um, this is not a reason to raise money. Um, but I I do think that it happens. Um, but I I do think that it happens. Um, but I I do think that it happens. Um, the fourth reason to to to raise capital the fourth reason to to to raise capital the fourth reason to to to raise capital is to reward your employees, right? It's is to reward your employees, right? It's is to reward your employees, right? It's a you're on a very long journey and you a you're on a very long journey and you a you're on a very long journey and you want to work with the best people in the want to work with the best people in the want to work with the best people in the world and by definition there's not that world and by definition there's not that world and by definition there's not that many best people in the world. So, you many best people in the world. So, you many best people in the world. So, you want to reward them. Um, that was the want to reward them. Um, that was the want to reward them. Um, that was the reason that we took more capital in reason that we took more capital in reason that we took more capital in December um was to allow the employees

December um was to allow the employees December um was to allow the employees to liquidate as some of their equity um to liquidate as some of their equity um to liquidate as some of their equity um instead of waiting for some like event instead of waiting for some like event instead of waiting for some like event like an IPO or something like further like an IPO or something like further like an IPO or something like further out. Um the fifth reason to raise is for out. Um the fifth reason to raise is for out. Um the fifth reason to raise is for a strategic partnership. There are a strategic partnership. There are a strategic partnership. There are strategic partnerships that have been strategic partnerships that have been strategic partnerships that have been made in this in this city that have made made in this in this city that have made made in this in this city that have made companies. Um and um the six reason to companies. Um and um the six reason to companies. Um and um the six reason to raise would be do doing M&A or or raise would be do doing M&A or or raise would be do doing M&A or or something like that. But it's like you something like that. But it's like you something like that. But it's like you have to be very honest about what reason have to be very honest about what reason have to be very honest about what reason you are raising in those six. First you are raising in those six. First you are raising in those six. First reason we raised was one and second reason we raised was one and second reason we raised was one and second reason we raised was four. Um so which reason we raised was four. Um so which reason we raised was four. Um so which which ones? The first reason to raise which ones? The first reason to raise which ones? The first reason to raise was R&D. R&D and the second reason was was R&D. R&D and the second reason was was R&D. R&D and the second reason was >> um to provide liquidity to the employees >> um to provide liquidity to the employees >> um to provide liquidity to the employees >> employees. Yep. >> employees. Yep. >> employees. Yep. >> I I think it's a it's a nice and healthy >> I I think it's a it's a nice and healthy >> I I think it's a it's a nice and healthy way and I think yeah the ego part we way and I think yeah the ego part we way and I think yeah the ego part we don't talk about and the identity and don't talk about and the identity and don't talk about and the identity and especially the closer you are to to tech especially the closer you are to to tech especially the closer you are to to tech ecosystems where a lot of people are ecosystems where a lot of people are ecosystems where a lot of people are raising it it will be part of it. As raising it it will be part of it. As raising it it will be part of it. As closing I I wanted to ask you about the closing I I wanted to ask you about the closing I I wanted to ask you about the way you have a remote culture these way you have a remote culture these way you have a remote culture these days. I'm seeing it especially for days. I'm seeing it especially for days. I'm seeing it especially for companies that do anything with AI, may companies that do anything with AI, may companies that do anything with AI, may that be building AI infra or or or just that be building AI infra or or or just that be building AI infra or or or just AI products. A lot of them prefer in AI products. A lot of them prefer in AI products. A lot of them prefer in person having a HQ often times in SF or person having a HQ often times in SF or person having a HQ often times in SF or wherever your headquarters may that be wherever your headquarters may that be wherever your headquarters may that be London or somewhere else because you London or somewhere else because you London or somewhere else because you often these companies often find that often these companies often find that often these companies often find that they have faster iteration. Uh it's just they have faster iteration. Uh it's just they have faster iteration. Uh it's just fewer layers cut in between and of fewer layers cut in between and of fewer layers cut in between and of course speed is is very very important.

course speed is is very very important. course speed is is very very important. You have started full remote and you're You have started full remote and you're You have started full remote and you're still full remote. how is it working? still full remote. how is it working? still full remote. how is it working? Uh, and what kind of quirks or like or Uh, and what kind of quirks or like or Uh, and what kind of quirks or like or turbo ways have you found to to make turbo ways have you found to to make turbo ways have you found to to make this work better? this work better? this work better? >> Yeah, I think so. The the company >> Yeah, I think so. The the company >> Yeah, I think so. The the company started in in 23. So, sort of like on started in in 23. So, sort of like on started in in 23. So, sort of like on the on the on the cusp of COVID where a the on the on the cusp of COVID where a the on the on the cusp of COVID where a lot of companies were just remote. Um, lot of companies were just remote. Um, lot of companies were just remote. Um, the Shopify infra was remote since the the Shopify infra was remote since the the Shopify infra was remote since the very um very beginning because it was very um very beginning because it was very um very beginning because it was very difficult to get them all to move very difficult to get them all to move very difficult to get them all to move to Ottawa. Um, and to Ottawa. Um, and to Ottawa. Um, and so it was natural to me. It's like, so it was natural to me. It's like, so it was natural to me. It's like, okay, I think there's kind of maybe two okay, I think there's kind of maybe two okay, I think there's kind of maybe two cities where you can build a database cities where you can build a database cities where you can build a database company fast, and that's San Francisco company fast, and that's San Francisco company fast, and that's San Francisco and and and maybe New York. There are and and and maybe New York. There are and and and maybe New York. There are maybe other cities, right? But that's maybe other cities, right? But that's maybe other cities, right? But that's like kind of where it's been done. like kind of where it's been done. like kind of where it's been done. >> Yeah. >> Yeah. >> Yeah. >> And so if you don't want to do that, I >> And so if you don't want to do that, I >> And so if you don't want to do that, I think you have to go all in on on on think you have to go all in on on on think you have to go all in on on on some distributed model. some distributed model. some distributed model. >> And so we've tried to figure out what >> And so we've tried to figure out what >> And so we've tried to figure out what does that distributed model mean for does that distributed model mean for does that distributed model mean for Turppuffer? It doesn't mean the absence Turppuffer? It doesn't mean the absence Turppuffer? It doesn't mean the absence of in person. we get everyone together of in person. we get everyone together of in person. we get everyone together twice a year in in some in in some twice a year in in some in in some twice a year in in some in in some location. Uh earlier this year we were location. Uh earlier this year we were location. Uh earlier this year we were in in B, right? And then we were in in in B, right? And then we were in in in B, right? And then we were in Mexico City and so on. So it's like Mexico City and so on. So it's like Mexico City and so on. So it's like that's that's not that uncommon. Um but that's that's not that uncommon. Um but that's that's not that uncommon. Um but one of the things that we we we've been one of the things that we we we've been one of the things that we we we've been trying to do is we have this concept trying to do is we have this concept trying to do is we have this concept called campfires. And the concept of the called campfires. And the concept of the called campfires. And the concept of the campfire is that when a couple of people campfire is that when a couple of people campfire is that when a couple of people just sort of randomly congregate in a just sort of randomly congregate in a just sort of randomly congregate in a place, you call it a campfire and you place, you call it a campfire and you place, you call it a campfire and you encourage as many people as you want to encourage as many people as you want to encourage as many people as you want to come and join. So, for example, this come and join. So, for example, this come and join. So, for example, this week is a Turbo Puffer campfire in San week is a Turbo Puffer campfire in San week is a Turbo Puffer campfire in San Francisco because I'm here for this Francisco because I'm here for this Francisco because I'm here for this conference and a bunch of other things.

conference and a bunch of other things. conference and a bunch of other things. And so, everyone is invited to come. And so, everyone is invited to come. And so, everyone is invited to come. Like, we're going to go meet customers, Like, we're going to go meet customers, Like, we're going to go meet customers, right? We're going to put on dinners for right? We're going to put on dinners for right? We're going to put on dinners for our customers and things like that. And our customers and things like that. And our customers and things like that. And we just make a thing out of it and and we just make a thing out of it and and we just make a thing out of it and and spend time together. And uh we encourage spend time together. And uh we encourage spend time together. And uh we encourage everyone to come. We've also gone to the everyone to come. We've also gone to the everyone to come. We've also gone to the extent now of um we want to encourage extent now of um we want to encourage extent now of um we want to encourage that, but not everyone not everyone that, but not everyone not everyone that, but not everyone not everyone needs to go to the campfire all the needs to go to the campfire all the needs to go to the campfire all the time. Some people just want to, you time. Some people just want to, you time. Some people just want to, you know, lock in and hacks into tent and know, lock in and hacks into tent and know, lock in and hacks into tent and that's great. We have people that just that's great. We have people that just that's great. We have people that just make it to the off sites twice a year make it to the off sites twice a year make it to the off sites twice a year and otherwise they're home, they're with and otherwise they're home, they're with and otherwise they're home, they're with their families and they don't they don't their families and they don't they don't their families and they don't they don't spend time on an airplane. Um, spend time on an airplane. Um, spend time on an airplane. Um, fantastic. Like that is completely fantastic. Like that is completely fantastic. Like that is completely compatible with this model. And there compatible with this model. And there compatible with this model. And there are other people at the company who are are other people at the company who are are other people at the company who are on a plane probably every two weeks. Um, on a plane probably every two weeks. Um, on a plane probably every two weeks. Um, we had someone the other day where they we had someone the other day where they we had someone the other day where they saw a campfire happening in New York and saw a campfire happening in New York and saw a campfire happening in New York and everyone was dialing in from a meeting everyone was dialing in from a meeting everyone was dialing in from a meeting room in New York and she had so much room in New York and she had so much room in New York and she had so much FOMO that she took an Uber straight to FOMO that she took an Uber straight to FOMO that she took an Uber straight to the airport in Ottawa and flew to the airport in Ottawa and flew to the airport in Ottawa and flew to [laughter] flew to New York to hang out [laughter] flew to New York to hang out [laughter] flew to New York to hang out with the team, right? And I think that's with the team, right? And I think that's with the team, right? And I think that's fantastic. Um and we've also introduced fantastic. Um and we've also introduced fantastic. Um and we've also introduced these things where um if you if you uh these things where um if you if you uh these things where um if you if you uh if you do a current conference talk or a if you do a current conference talk or a if you do a current conference talk or a blog post or something like that at blog post or something like that at blog post or something like that at Turbop or something a bit Turbop or something a bit Turbop or something a bit extracurricular, we give you a turbo extracurricular, we give you a turbo extracurricular, we give you a turbo credit and a turbo credit allows you to credit and a turbo credit allows you to credit and a turbo credit allows you to upgrade your next flight to business upgrade your next flight to business upgrade your next flight to business class which again encourages spending class which again encourages spending class which again encourages spending time together with the team. Um and now time together with the team. Um and now time together with the team. Um and now I mean turbo credits are probably going I mean turbo credits are probably going I mean turbo credits are probably going to take on a life of their own. Someone to take on a life of their own. Someone to take on a life of their own. Someone was talking about doing a central bank was talking about doing a central bank was talking about doing a central bank and doing interest rates on the turbo and doing interest rates on the turbo and doing interest rates on the turbo credits. um and doing a betting market credits. um and doing a betting market credits. um and doing a betting market on the turbo credits. And so like this on the turbo credits. And so like this on the turbo credits. And so like this might take on its life on its own. Um

might take on its life on its own. Um might take on its life on its own. Um and uh you you know if you um if you're and uh you you know if you um if you're and uh you you know if you um if you're at a conference like this, there's some at a conference like this, there's some at a conference like this, there's some of the our engineers here who are just of the our engineers here who are just of the our engineers here who are just want to interact with customers and be want to interact with customers and be want to interact with customers and be on like and standing on a like expo on like and standing on a like expo on like and standing on a like expo floor all day is quite taxing. And so if floor all day is quite taxing. And so if floor all day is quite taxing. And so if you do that for two days because you you do that for two days because you you do that for two days because you want to do it, oh, you get a turbo want to do it, oh, you get a turbo want to do it, oh, you get a turbo credit, right? And so it's just like credit, right? And so it's just like credit, right? And so it's just like these fun little things that we try to these fun little things that we try to these fun little things that we try to do to to to encourage people to meet if do to to to encourage people to meet if do to to to encourage people to meet if they want to meet Thank you. Well, in they want to meet Thank you. Well, in they want to meet Thank you. Well, in this session, uh, what I found very this session, uh, what I found very this session, uh, what I found very interesting is Turbopuffer is a so many interesting is Turbopuffer is a so many interesting is Turbopuffer is a so many AI companies are using you as an AI companies are using you as an AI companies are using you as an infrastructure layer, but in this infrastructure layer, but in this infrastructure layer, but in this conversation, we managed to talk very conversation, we managed to talk very conversation, we managed to talk very little about AI and a lot more about little about AI and a lot more about little about AI and a lot more about engineering principles, pushing, being engineering principles, pushing, being engineering principles, pushing, being curious, and the human connection, how curious, and the human connection, how curious, and the human connection, how important it is for people to work important it is for people to work important it is for people to work together, to trust each other. So, just together, to trust each other. So, just together, to trust each other. So, just thank you very much for that. So, let's thank you very much for that. So, let's thank you very much for that. So, let's give a big round of applause for Simon. give a big round of applause for Simon. give a big round of applause for Simon. >> Thank you so much. This is great. Thank >> Thank you so much. This is great. Thank >> Thank you so much. This is great. Thank you. you. you. Are you