← Back
AI Engineer August 9, 2026 20m

Velocity Sickness: What Happens When Your Whole Team Gets 10x Faster — Matt Dailey, Ref.

Read full transcript 18 segments
  1. Thank you so much for coming later in Thank you so much for coming later in the week here. the week here. the week here. Uh Uh Uh I'm Matt. I'm Matt. I'm Matt. I'm the CEO and founder of Ref. I'm the CEO and founder of Ref. I'm the CEO and founder of Ref. The problem we work on at Ref is one you The problem we work on at Ref is one you The problem we work on at Ref is one you might be familiar with where individual might be familiar with where individual might be familiar with where individual engineers are going really fast with AI, engineers are going really fast with AI, engineers are going really fast with AI, but the team as a whole is not. And but the team as a whole is not. And but the team as a whole is not. And we're working to help close that gap. we're working to help close that gap. we're working to help close that gap. What I'm going to be talking about today What I'm going to be talking about today What I'm going to be talking about today is that is what happens when your whole is that is what happens when your whole is that is what happens when your whole team gets 10 times faster. team gets 10 times faster. team gets 10 times faster. All the things that don't go well. All the things that don't go well. All the things that don't go well. And my goal is to give you a blueprint And my goal is to give you a blueprint And my goal is to give you a blueprint for how to get get through those issues. for how to get get through those issues. for how to get get through those issues. Uh and have the whole team move Uh and have the whole team move Uh and have the whole team move faster together. faster together. faster together. The way I want to accomplish that is The way I want to accomplish that is The way I want to accomplish that is first first first talk about those issues, define some talk about those issues, define some talk about those issues, define some terms. terms. terms. Then we'll talk about how did we get Then we'll talk about how did we get Then we'll talk about how did we get here? here? here? Where are we at now? Where are we at now? Where are we at now? Um and we'll use that to Um and we'll use that to Um and we'll use that to triangulate on some solutions. We'll triangulate on some solutions. We'll triangulate on some solutions. We'll start high level and we'll work down to start high level and we'll work down to start high level and we'll work down to get more and more practical until we get more and more practical until we get more and more practical until we leave with a couple of things you can leave with a couple of things you can leave with a couple of things you can leave here and do immediately. leave here and do immediately. leave here and do immediately. Um how's that sound? I need thumbs up.

  2. Um how's that sound? I need thumbs up. Um how's that sound? I need thumbs up. All right, awesome. Great, that was All right, awesome. Great, that was All right, awesome. Great, that was great. Uh great. Uh great. Uh All right, so let's get into it. All right, so let's get into it. All right, so let's get into it. These are some of the problems you might These are some of the problems you might These are some of the problems you might be experiencing right now. be experiencing right now. be experiencing right now. Uh these are these are problems that Uh these are these are problems that Uh these are these are problems that affect individuals and are actually affect individuals and are actually affect individuals and are actually magnified magnified magnified at the team level once the entire team at the team level once the entire team at the team level once the entire team starts experiencing them. starts experiencing them. starts experiencing them. The first one, The first one, The first one, too many PRs to merge. This is the like too many PRs to merge. This is the like too many PRs to merge. This is the like classic first problem you hit when you classic first problem you hit when you classic first problem you hit when you start adopting AI as an engineer. You're start adopting AI as an engineer. You're start adopting AI as an engineer. You're like, "Great, I'm shipping stuff." And like, "Great, I'm shipping stuff." And like, "Great, I'm shipping stuff." And you push up those PRs and you're like, you push up those PRs and you're like, you push up those PRs and you're like, "There's no way I can merge all these." "There's no way I can merge all these." "There's no way I can merge all these." Uh and this is obviously magnified when Uh and this is obviously magnified when Uh and this is obviously magnified when the whole team does this. the whole team does this. the whole team does this. Merge conflicts, merge queue breaks Merge conflicts, merge queue breaks Merge conflicts, merge queue breaks down, down, down, things get bad. things get bad. things get bad. The second problem is that you're moving The second problem is that you're moving The second problem is that you're moving in many directions at once. in many directions at once. in many directions at once. This is also both individual and a team This is also both individual and a team This is also both individual and a team problem. Individually, this is where you problem. Individually, this is where you problem. Individually, this is where you have a bunch of agents doing different have a bunch of agents doing different have a bunch of agents doing different things. You're trying to remember who's things. You're trying to remember who's things. You're trying to remember who's doing what doing what doing what and your brain gets fried. and your brain gets fried. and your brain gets fried. At the organizational level, this is At the organizational level, this is At the organizational level, this is your engineers your engineers your engineers picking up things and running in a picking up things and running in a picking up things and running in a certain direction. Another engineer is certain direction. Another engineer is certain direction. Another engineer is running over here.

  3. running over here. running over here. Maybe some are bumping into each other Maybe some are bumping into each other Maybe some are bumping into each other and you're you're not moving cohesively and you're you're not moving cohesively and you're you're not moving cohesively with focus because you're just sprinting with focus because you're just sprinting with focus because you're just sprinting in all sorts of directions. in all sorts of directions. in all sorts of directions. The third problem you might be The third problem you might be The third problem you might be experiencing is declaring agent experiencing is declaring agent experiencing is declaring agent bankruptcy. This is a a common pattern I bankruptcy. This is a a common pattern I bankruptcy. This is a a common pattern I see engineers get into where see engineers get into where see engineers get into where you know, you're cranking. You have like you know, you're cranking. You have like you know, you're cranking. You have like your 12 terminals open. You're like at your 12 terminals open. You're like at your 12 terminals open. You're like at the end of the day you're like, "Yeah, I the end of the day you're like, "Yeah, I the end of the day you're like, "Yeah, I did a lot of work." You step away from did a lot of work." You step away from did a lot of work." You step away from your laptop. your laptop. your laptop. Uh Uh Uh spend time with your friends and family. spend time with your friends and family. spend time with your friends and family. Next morning you come back and it's like Next morning you come back and it's like Next morning you come back and it's like walking into just a room of strangers. walking into just a room of strangers. walking into just a room of strangers. Like, who are these people? What are Like, who are these people? What are Like, who are these people? What are they doing here? But like they doing here? But like they doing here? But like no problem. They're agents so you just no problem. They're agents so you just no problem. They're agents so you just like like like get rid of them and start over again. get rid of them and start over again. get rid of them and start over again. The problem though is you know, it feels The problem though is you know, it feels The problem though is you know, it feels like you're doing a lot of work but like you're doing a lot of work but like you're doing a lot of work but you're doing the same work you're doing the same work you're doing the same work and you're spending tokens twice. You're and you're spending tokens twice. You're and you're spending tokens twice. You're doing the same problems over over again. doing the same problems over over again. doing the same problems over over again. And if you think about that And if you think about that And if you think about that organizationally, that's your team not organizationally, that's your team not organizationally, that's your team not being efficient with both their time and being efficient with both their time and being efficient with both their time and their their token resources. Problem four though is actually the most Problem four though is actually the most important one.

  4. important one. important one. It's critical decisions being made by It's critical decisions being made by It's critical decisions being made by agents. Once you have agents agents. Once you have agents agents. Once you have agents doing a lot of your work, doing a lot of your work, doing a lot of your work, if if you as an engineer are letting an if if you as an engineer are letting an if if you as an engineer are letting an agent make a critical decision, you are agent make a critical decision, you are agent make a critical decision, you are seeding control of your code. You are no seeding control of your code. You are no seeding control of your code. You are no longer the owner of that code. The agent longer the owner of that code. The agent longer the owner of that code. The agent is. is. is. And if you imagine that at scale at your And if you imagine that at scale at your And if you imagine that at scale at your company, if the engineers across your company, if the engineers across your company, if the engineers across your team are you know, giving up ownership team are you know, giving up ownership team are you know, giving up ownership of the code, you no longer own the of the code, you no longer own the of the code, you no longer own the product. These are a bunch of problems you might These are a bunch of problems you might be experiencing at different degrees be experiencing at different degrees be experiencing at different degrees at different days. Uh at different days. Uh at different days. Uh The way I like bundle it up is into this The way I like bundle it up is into this The way I like bundle it up is into this term velocity sickness. term velocity sickness. term velocity sickness. This is uh This is uh This is uh the stress caused by sudden output the stress caused by sudden output the stress caused by sudden output increases thanks to AI. Um it affects increases thanks to AI. Um it affects increases thanks to AI. Um it affects individuals or teams. individuals or teams. individuals or teams. Um and the result is output without Um and the result is output without Um and the result is output without impact. So, this is that feeling of like impact. So, this is that feeling of like impact. So, this is that feeling of like we're moving really fast. This should we're moving really fast. This should we're moving really fast. This should feel great. This should feel awesome, feel great. This should feel awesome, feel great. This should feel awesome, but for some reason it doesn't feel but for some reason it doesn't feel but for some reason it doesn't feel awesome. We're not having the like the awesome. We're not having the like the awesome. We're not having the like the things you expect to be happening are things you expect to be happening are things you expect to be happening are not happening. Um despite this feeling not happening. Um despite this feeling not happening. Um despite this feeling of being so productive.

  5. So, that's to define that term, but I So, that's to define that term, but I want to tell uh a little story. want to tell uh a little story. want to tell uh a little story. Uh Uh Uh my talk is happening now. my talk is happening now. my talk is happening now. Uh I want to tell a little story I want to tell a little story about somebody like to make this very about somebody like to make this very about somebody like to make this very real. real. real. Um This is a story about somebody uh who's This is a story about somebody uh who's actually not an engineer, but I think it actually not an engineer, but I think it actually not an engineer, but I think it parallels our engineering workflows a parallels our engineering workflows a parallels our engineering workflows a lot. I Part of my job is I get to go, lot. I Part of my job is I get to go, lot. I Part of my job is I get to go, find, and talk to people who are really find, and talk to people who are really find, and talk to people who are really pushing the forefront and try and learn pushing the forefront and try and learn pushing the forefront and try and learn from them. And I love that part of it. from them. And I love that part of it. from them. And I love that part of it. And so, this is somebody and I'm I'm And so, this is somebody and I'm I'm And so, this is somebody and I'm I'm talking through their identical talking through their identical talking through their identical workflows. workflows. workflows. Um and this is somebody who writes a Um and this is somebody who writes a Um and this is somebody who writes a newsletter, and newsletter, and newsletter, and their workflow is around how do I I have their workflow is around how do I I have their workflow is around how do I I have my ideas and then they're doing research my ideas and then they're doing research my ideas and then they're doing research and exploration and then uh like and exploration and then uh like and exploration and then uh like cohesion between those ideas and cohesion between those ideas and cohesion between those ideas and editorial to make sure it's in their editorial to make sure it's in their editorial to make sure it's in their voice. And they're they're walking me voice. And they're they're walking me voice. And they're they're walking me through their system for how they they through their system for how they they through their system for how they they manage this whole pipeline um and really manage this whole pipeline um and really manage this whole pipeline um and really scale out their scale out their scale out their their efficacy. And this is like very their efficacy. And this is like very their efficacy. And this is like very very much not slop. This is somebody who very much not slop. This is somebody who very much not slop. This is somebody who has is using agents to amplify their own has is using agents to amplify their own has is using agents to amplify their own voice uh in a way that's like very voice uh in a way that's like very voice uh in a way that's like very impressive. I'm like, "Wow, that's that impressive. I'm like, "Wow, that's that impressive. I'm like, "Wow, that's that is so cool. I'm I'm like is so cool. I'm I'm like is so cool. I'm I'm like I'm soaking it in." Uh I'm soaking it in." Uh I'm soaking it in." Uh And then they're like, "Yeah, I'm And then they're like, "Yeah, I'm And then they're like, "Yeah, I'm basically writing a book every week."

  6. basically writing a book every week." basically writing a book every week." And I'm like, "Oh, And I'm like, "Oh, And I'm like, "Oh, okay. Is like is your audience reading a okay. Is like is your audience reading a okay. Is like is your audience reading a book every week?" book every week?" book every week?" And they're like, "No, they're probably And they're like, "No, they're probably And they're like, "No, they're probably not." Right? They're not. They're not." Right? They're not. They're not." Right? They're not. They're they're This is a person who's they're This is a person who's they're This is a person who's writing a lot, but those pages that writing a lot, but those pages that writing a lot, but those pages that they're writing they're writing they're writing are going unread. Um are going unread. Um are going unread. Um I think that's part of this experience I think that's part of this experience I think that's part of this experience of velocity sickness is we're we're of velocity sickness is we're we're of velocity sickness is we're we're building things that are not building things that are not building things that are not mattering for the people we want them to mattering for the people we want them to mattering for the people we want them to matter for, the people we're trying to matter for, the people we're trying to matter for, the people we're trying to reach. Uh So, what we Instead of unread reach. Uh So, what we Instead of unread reach. Uh So, what we Instead of unread pages, pages, pages, um what we want is we want to write um what we want is we want to write um what we want is we want to write words that matter. We want to write words that matter. We want to write words that matter. We want to write words that connect with people. words that connect with people. words that connect with people. Um and the parallel to us as like Um and the parallel to us as like Um and the parallel to us as like software builders and product builders software builders and product builders software builders and product builders is is that is is that is is that we want to write we want to build we want to write we want to build we want to write we want to build products that connect with people. We products that connect with people. We products that connect with people. We want We want to build products that want We want to build products that want We want to build products that change the way people live and work and change the way people live and work and change the way people live and work and make their lives better. And we have make their lives better. And we have make their lives better. And we have more ability to do that than ever with more ability to do that than ever with more ability to do that than ever with AI. AI. AI. But we have this like feeling of But we have this like feeling of But we have this like feeling of velocity sickness where we're It feels velocity sickness where we're It feels velocity sickness where we're It feels like we should be doing that, but it's like we should be doing that, but it's like we should be doing that, but it's not quite landing. not quite landing. not quite landing. So, let's talk about how we got here. So, let's talk about how we got here. So, let's talk about how we got here. Um Um Um This is what the software engineering This is what the software engineering This is what the software engineering process used to look like before AI.

  7. process used to look like before AI. process used to look like before AI. We'd do some planning up front. We'd do some planning up front. We'd do some planning up front. We'd sit down and build. We'd implement. We'd sit down and build. We'd implement. We'd sit down and build. We'd implement. It'd be It'd be iterative. We'd be It'd be It'd be iterative. We'd be It'd be It'd be iterative. We'd be exploring. But largely we're sitting exploring. But largely we're sitting exploring. But largely we're sitting down building down building down building in isolation, implementing something. in isolation, implementing something. in isolation, implementing something. And at the end we sort of polish it up And at the end we sort of polish it up And at the end we sort of polish it up and ship it out the door. And this was and ship it out the door. And this was and ship it out the door. And this was great. We all knew how to do this. We great. We all knew how to do this. We great. We all knew how to do this. We had a lot of systems for this. Uh had a lot of systems for this. Uh had a lot of systems for this. Uh And it was good. And then AI came along. And it was good. And then AI came along. And it was good. And then AI came along. Um Um Um But But But at this time our our tools were built at this time our our tools were built at this time our our tools were built for this. Like all all our history of for this. Like all all our history of for this. Like all all our history of coding tools were built for this style coding tools were built for this style coding tools were built for this style of work. of work. of work. Um our IDE, our workhorse, um it was Um our IDE, our workhorse, um it was Um our IDE, our workhorse, um it was built for implementation and polish to built for implementation and polish to built for implementation and polish to be done by an individual, to be heads be done by an individual, to be heads be done by an individual, to be heads down building as a software engineer down building as a software engineer down building as a software engineer writing code. writing code. writing code. Um That's what That's a tool built for Um That's what That's a tool built for Um That's what That's a tool built for how we used to work. how we used to work. how we used to work. Um so, let's look at how how we work Um so, let's look at how how we work Um so, let's look at how how we work now. now. now. Our work looks a lot more like this. Our work looks a lot more like this. Our work looks a lot more like this. Where you do some planning up front. You Where you do some planning up front. You Where you do some planning up front. You start thinking about what am I trying to start thinking about what am I trying to start thinking about what am I trying to do here? You take this idea in your head do here? You take this idea in your head do here? You take this idea in your head and start to flesh it out. and start to flesh it out. and start to flesh it out. Um At some point an agent takes that Um At some point an agent takes that Um At some point an agent takes that idea and like implements it. This is idea and like implements it. This is idea and like implements it. This is like arguably should not even be on this like arguably should not even be on this like arguably should not even be on this slide cuz it's not our human work slide cuz it's not our human work slide cuz it's not our human work anymore. It's done by the agent. Um anymore. It's done by the agent. Um anymore. It's done by the agent. Um and then in the end we do some polishing and then in the end we do some polishing and then in the end we do some polishing where we take back that thing the agent where we take back that thing the agent where we take back that thing the agent has made for us and we like hold it in has made for us and we like hold it in has made for us and we like hold it in our hands and we say, is this is this our hands and we say, is this is this our hands and we say, is this is this what I wanted? Um what I wanted? Um what I wanted? Um This is a very different shape of work.

  8. This is a very different shape of work. This is a very different shape of work. Right? We're no longer doing this like Right? We're no longer doing this like Right? We're no longer doing this like heads down building. heads down building. heads down building. These are the two uh These are the two uh These are the two uh creative and collaborative parts of our creative and collaborative parts of our creative and collaborative parts of our work as engineers. This is where we like work as engineers. This is where we like work as engineers. This is where we like express our craft as an engineer. And express our craft as an engineer. And express our craft as an engineer. And the one that's most different is the the the one that's most different is the the the one that's most different is the the planning stage. So, I want to zoom in on planning stage. So, I want to zoom in on planning stage. So, I want to zoom in on that just a little bit. that just a little bit. that just a little bit. Uh that's the like exploratory, Uh that's the like exploratory, Uh that's the like exploratory, creative, collaborative part where we're creative, collaborative part where we're creative, collaborative part where we're we're thinking about like I have this we're thinking about like I have this we're thinking about like I have this vague idea. vague idea. vague idea. I have this complex system. I need to I have this complex system. I need to I have this complex system. I need to understand the contours of this system, understand the contours of this system, understand the contours of this system, apply this idea, and like apply this idea, and like apply this idea, and like really understand it. And I need to pull really understand it. And I need to pull really understand it. And I need to pull out what's relevant and express my taste out what's relevant and express my taste out what's relevant and express my taste as an engineer. As to like, where do I as an engineer. As to like, where do I as an engineer. As to like, where do I want this system to go? want this system to go? want this system to go? Um Um Um This is the kind of work that uh This is the kind of work that uh This is the kind of work that uh that's creative and and needs to be done that's creative and and needs to be done that's creative and and needs to be done together. I think the the shape of together. I think the the shape of together. I think the the shape of engineering teams and product teams is engineering teams and product teams is engineering teams and product teams is changing with AI. changing with AI. changing with AI. But we're always going to have this But we're always going to have this But we're always going to have this thing where we have a group of people thing where we have a group of people thing where we have a group of people responsible for managing a complex responsible for managing a complex responsible for managing a complex system and it's sort of deciding the system and it's sort of deciding the system and it's sort of deciding the future of that system. And that's this future of that system. And that's this future of that system. And that's this planning work. And so, we're planning work. And so, we're planning work. And so, we're we're starting to see some of the hints we're starting to see some of the hints we're starting to see some of the hints of like, okay, we had tools built for a of like, okay, we had tools built for a of like, okay, we had tools built for a certain type of work. Our work looks certain type of work. Our work looks certain type of work. Our work looks different now.

  9. different now. different now. Um Let's start to see think about that a Um Let's start to see think about that a Um Let's start to see think about that a little bit more. little bit more. little bit more. So, uh we used to have the IDE as our So, uh we used to have the IDE as our So, uh we used to have the IDE as our main tool. It was used for main tool. It was used for main tool. It was used for implementation. implementation. implementation. But now we're doing this different kind But now we're doing this different kind But now we're doing this different kind of work. So, at the very high level, we of work. So, at the very high level, we of work. So, at the very high level, we can start to think about solutions like, can start to think about solutions like, can start to think about solutions like, okay, we have a tool built for this type okay, we have a tool built for this type okay, we have a tool built for this type of work. We have a new type of work. of work. We have a new type of work. of work. We have a new type of work. Let's think about Let's think about Let's think about you know, what this new layer of our you know, what this new layer of our you know, what this new layer of our work is. This This is the decision work is. This This is the decision work is. This This is the decision layer. This is where we're thinking layer. This is where we're thinking layer. This is where we're thinking through through through what are the key decisions, doing that what are the key decisions, doing that what are the key decisions, doing that like craft of engineering, um and like craft of engineering, um and like craft of engineering, um and expressing our taste as engineers, expressing our taste as engineers, expressing our taste as engineers, ultimately a different thing than ultimately a different thing than ultimately a different thing than implementation. implementation. implementation. And it And it And it it's a different gear as an engineer. it's a different gear as an engineer. it's a different gear as an engineer. The skill now is what gear am I in? Am I The skill now is what gear am I in? Am I The skill now is what gear am I in? Am I using the appropriate tools for the gear using the appropriate tools for the gear using the appropriate tools for the gear that I'm what I'm trying to accomplish that I'm what I'm trying to accomplish that I'm what I'm trying to accomplish right now. So let's think about what a tool built So let's think about what a tool built for the decision layer would look like. It'd be a tool built for docs It'd be a tool built for docs and not chat. And there's that's like a and not chat. And there's that's like a and not chat. And there's that's like a short sentence. There's a lot to unpack short sentence. There's a lot to unpack short sentence. There's a lot to unpack here though. So I'm going to spend a lot here though. So I'm going to spend a lot here though. So I'm going to spend a lot of time on this slide. of time on this slide. of time on this slide. The problem with chats The problem with chats The problem with chats is that they are is that they are is that they are the relic of building for the relic of building for the relic of building for implementation. So they're implementation. So they're implementation. So they're they're default isolated and ephemeral they're default isolated and ephemeral they're default isolated and ephemeral and and brain off.

  10. and and brain off. and and brain off. They're they're made to build things and They're they're made to build things and They're they're made to build things and get stuff done. And that's not really get stuff done. And that's not really get stuff done. And that's not really the same type of work we're doing the same type of work we're doing the same type of work we're doing at the decision layer. We're doing this at the decision layer. We're doing this at the decision layer. We're doing this creative exploratory work. So being in creative exploratory work. So being in creative exploratory work. So being in this isolated environment where I'm this isolated environment where I'm this isolated environment where I'm working with an agent, maybe I start working with an agent, maybe I start working with an agent, maybe I start with my vague idea and I'm exploring it with my vague idea and I'm exploring it with my vague idea and I'm exploring it and asking questions, but decisions are and asking questions, but decisions are and asking questions, but decisions are being made in that like that chat that being made in that like that chat that being made in that like that chat that are are not shared with my team are are not shared with my team are are not shared with my team that are that are that are going to disappear as as long I'm going going to disappear as as long I'm going going to disappear as as long I'm going to to to result in some code being output where result in some code being output where result in some code being output where those important decisions are not being those important decisions are not being those important decisions are not being made clear and shared with the team. And made clear and shared with the team. And made clear and shared with the team. And you're also in this mode where the agent you're also in this mode where the agent you're also in this mode where the agent saying saying saying Okay, this is what I want to do. Is is Okay, this is what I want to do. Is is Okay, this is what I want to do. Is is that okay? Let's go. Or or sometimes that okay? Let's go. Or or sometimes that okay? Let's go. Or or sometimes it'll say it'll ask you a question and it'll say it'll ask you a question and it'll say it'll ask you a question and it'll be like it'll be like it'll be like you know, this is the recommended option you know, this is the recommended option you know, this is the recommended option and then you're like, great. I don't and then you're like, great. I don't and then you're like, great. I don't even think about this. I'll just hit even think about this. I'll just hit even think about this. I'll just hit that one and we keep going. Docs start to solve this problem. So Docs start to solve this problem. So when you're if you center your work when you're if you center your work when you're if you center your work around around around working in docs they're meant to be bring forward the they're meant to be bring forward the key decisions. This is this is what our key decisions. This is this is what our key decisions. This is this is what our work is now. Our work now is work is now. Our work now is work is now. Our work now is figure out what decisions matter and figure out what decisions matter and figure out what decisions matter and then make those decisions then make those decisions then make those decisions and then get out of the way while the and then get out of the way while the and then get out of the way while the agents fill in the rest.

  11. Working in Docs is the the classic way Working in Docs is the the classic way we would create alignment. we would create alignment. we would create alignment. If you were If you think back to being a If you were If you think back to being a If you were If you think back to being a manager or manager or manager or a lead on a team before AI, if your team a lead on a team before AI, if your team a lead on a team before AI, if your team was having struggling with alignment, was having struggling with alignment, was having struggling with alignment, you would not tell them like, "Let's go you would not tell them like, "Let's go you would not tell them like, "Let's go all work in Slack DMs. Let's like go all work in Slack DMs. Let's like go all work in Slack DMs. Let's like go direct message each other." You'd say, direct message each other." You'd say, direct message each other." You'd say, "Let's "Let's "Let's like bring forward key decisions, align like bring forward key decisions, align like bring forward key decisions, align on them, spend time on these, like find on them, spend time on these, like find on them, spend time on these, like find a way that we can really spend time a way that we can really spend time a way that we can really spend time getting our decisions right." So, there are two things you might be So, there are two things you might be thinking looking at this that are not thinking looking at this that are not thinking looking at this that are not what I'm talking about here. The first what I'm talking about here. The first what I'm talking about here. The first one is plan mode. That's a great tool. one is plan mode. That's a great tool. one is plan mode. That's a great tool. The other one is like full-on factory The other one is like full-on factory The other one is like full-on factory spectrum and development. That's a great spectrum and development. That's a great spectrum and development. That's a great tool, too. I'm talking about something tool, too. I'm talking about something tool, too. I'm talking about something in the middle. in the middle. in the middle. Plan mode is great, but it's largely Plan mode is great, but it's largely Plan mode is great, but it's largely a a rich chat message where the agent is a a rich chat message where the agent is a a rich chat message where the agent is saying, "Hey, here's a like better saying, "Hey, here's a like better saying, "Hey, here's a like better visualization of what I'm trying to visualization of what I'm trying to visualization of what I'm trying to express to you." Yeah, that's great. express to you." Yeah, that's great. express to you." Yeah, that's great. But it's still in this isolated But it's still in this isolated But it's still in this isolated ephemeral environment. ephemeral environment. ephemeral environment. And what I'm suggesting is something And what I'm suggesting is something And what I'm suggesting is something more more durable, more more durable, more more durable, more shared, more long-lived that you more shared, more long-lived that you more shared, more long-lived that you and your team are spending time on.

  12. and your team are spending time on. and your team are spending time on. Similarly, the spectrum and development Similarly, the spectrum and development Similarly, the spectrum and development where we just define the behaviors, we where we just define the behaviors, we where we just define the behaviors, we operate at the like product level, operate at the like product level, operate at the like product level, is a little far away from the is a little far away from the is a little far away from the engineering reality. engineering reality. engineering reality. The engineering reality is that I need The engineering reality is that I need The engineering reality is that I need to understand my system and have a tool to understand my system and have a tool to understand my system and have a tool that that that like helps me understand that system and like helps me understand that system and like helps me understand that system and lay out those key decisions in a in a lay out those key decisions in a in a lay out those key decisions in a in a technical sense. The way I like to technical sense. The way I like to technical sense. The way I like to conceptualize this is as the portal to conceptualize this is as the portal to conceptualize this is as the portal to the software system, where you are like the software system, where you are like the software system, where you are like Tony Stark and you're like, "Show me Tony Stark and you're like, "Show me Tony Stark and you're like, "Show me what matters." And you're like, "I'm what matters." And you're like, "I'm what matters." And you're like, "I'm working on this. Pull out the bits that working on this. Pull out the bits that working on this. Pull out the bits that are relevant." And AI is amazing at are relevant." And AI is amazing at are relevant." And AI is amazing at finding things that are related to other finding things that are related to other finding things that are related to other things, helping you find what's what's things, helping you find what's what's things, helping you find what's what's relevant and lay them out on the table relevant and lay them out on the table relevant and lay them out on the table in front of you. Organize the pieces in in front of you. Organize the pieces in in front of you. Organize the pieces in the way you want to represent how you the way you want to represent how you the way you want to represent how you want the system to grow. But the big conceptual flip here is But the big conceptual flip here is actually that we're pulling out the actually that we're pulling out the actually that we're pulling out the state. So, when you're living and state. So, when you're living and state. So, when you're living and working in a working in a working in a a long-lived session with an agent, a long-lived session with an agent, a long-lived session with an agent, there's this implicit context being there's this implicit context being there's this implicit context being built up like over that work, and that's built up like over that work, and that's built up like over that work, and that's great. Um great. Um great. Um But there's you're also doing actions But there's you're also doing actions But there's you're also doing actions and it's it's not shared. What you want and it's it's not shared. What you want and it's it's not shared. What you want is to separate the is to separate the is to separate the the agent as the action and the doc as the agent as the action and the doc as the agent as the action and the doc as the state. And so, you can spawn new the state. And so, you can spawn new the state. And so, you can spawn new agents that have the same context or agents that have the same context or agents that have the same context or starting from the same place starting from the same place starting from the same place that are able to collaborate and work on that are able to collaborate and work on that are able to collaborate and work on the same the same the same same same same piece of context and state. You're piece of context and state. You're piece of context and state. You're ultimately doing context engineering in ultimately doing context engineering in ultimately doing context engineering in this doc, so that every agent is largely this doc, so that every agent is largely this doc, so that every agent is largely stateless and starts from this place um stateless and starts from this place um stateless and starts from this place um the same place. And what that gives you the same place. And what that gives you the same place. And what that gives you is your team and yourself can look into

  13. is your team and yourself can look into is your team and yourself can look into this and understand this and understand this and understand what actually is in here. What are the what actually is in here. What are the what actually is in here. What are the decisions that are being made um decisions that are being made um decisions that are being made um and have a a clear understanding of what and have a a clear understanding of what and have a a clear understanding of what the key decisions are. So, this is a a different way of working So, this is a a different way of working where your your core atom of your work where your your core atom of your work where your your core atom of your work is a doc rather than a chat. Um is a doc rather than a chat. Um is a doc rather than a chat. Um What happens when you actually start to What happens when you actually start to What happens when you actually start to implement this? implement this? implement this? Well, the first thing we see happen Well, the first thing we see happen Well, the first thing we see happen actually is that people start to plan actually is that people start to plan actually is that people start to plan and then not implement their plan. and then not implement their plan. and then not implement their plan. Uh and this is actually like a really Uh and this is actually like a really Uh and this is actually like a really good sign. good sign. good sign. Because what that means is they're Because what that means is they're Because what that means is they're thinking through ideas. They're saying, thinking through ideas. They're saying, thinking through ideas. They're saying, "I have this idea. Let me explore it. "I have this idea. Let me explore it. "I have this idea. Let me explore it. Let me start to flesh it out. Like I Let me start to flesh it out. Like I Let me start to flesh it out. Like I have this vague thought. Don't just give have this vague thought. Don't just give have this vague thought. Don't just give me some code. Don't go off and like me some code. Don't go off and like me some code. Don't go off and like build it for me, but let me help help me build it for me, but let me help help me build it for me, but let me help help me understand the idea that I'm talking understand the idea that I'm talking understand the idea that I'm talking about. Um about. Um about. Um And then they have a bunch of these and And then they have a bunch of these and And then they have a bunch of these and then some of them are getting built and then some of them are getting built and then some of them are getting built and some of them aren't. So, that means some of them aren't. So, that means some of them aren't. So, that means they're prioritizing the ideas that they're prioritizing the ideas that they're prioritizing the ideas that after they've explored them after they've explored them after they've explored them are the ones that are worth building and are the ones that are worth building and are the ones that are worth building and and going in this to the next step with. and going in this to the next step with. and going in this to the next step with. Um One way I like to frame this is that Um One way I like to frame this is that Um One way I like to frame this is that you're you're shifting from code you're you're shifting from code you're you're shifting from code velocity velocity velocity to idea velocity. So, to idea velocity. So, to idea velocity. So, going back to our going back to our going back to our our problem statement like how are we our problem statement like how are we our problem statement like how are we dealing with velocity sickness? The dealing with velocity sickness? The dealing with velocity sickness? The velocity sickness is we're shipping too velocity sickness is we're shipping too velocity sickness is we're shipping too much code that's not going anywhere. The much code that's not going anywhere. The much code that's not going anywhere. The solution solution solution is to shift that velocity to ideas so is to shift that velocity to ideas so is to shift that velocity to ideas so that rather than that rather than that rather than you know, getting stuck in prototype you know, getting stuck in prototype you know, getting stuck in prototype gravity where we we build something and gravity where we we build something and gravity where we we build something and we're so excited to just ship that thing we're so excited to just ship that thing we're so excited to just ship that thing and we're going down like one path of and we're going down like one path of and we're going down like one path of the idea maze,

  14. the idea maze, the idea maze, we can now like more more effectively we can now like more more effectively we can now like more more effectively explore that whole maze and find that explore that whole maze and find that explore that whole maze and find that gold that's around the corner. gold that's around the corner. gold that's around the corner. Um Um Um and and really and and really and and really impact the people we're trying to help. impact the people we're trying to help. impact the people we're trying to help. Um Um Um So, let's go back. Here are those So, let's go back. Here are those So, let's go back. Here are those problems, those velocity sickness problems, those velocity sickness problems, those velocity sickness problems I talked about the beginning. problems I talked about the beginning. problems I talked about the beginning. Let's see how this starts to address Let's see how this starts to address Let's see how this starts to address those. those. those. First, First, First, too many PRs. too many PRs. too many PRs. We've moved the the review point earlier We've moved the the review point earlier We've moved the the review point earlier into the process. So, we're aligning on into the process. So, we're aligning on into the process. So, we're aligning on the key decisions up front. the key decisions up front. the key decisions up front. That means the code review is easier That means the code review is easier That means the code review is easier because the hardest part of any code because the hardest part of any code because the hardest part of any code review is, review is, review is, you know, what actually matters here. you know, what actually matters here. you know, what actually matters here. That's the first step is like, "Hmm, That's the first step is like, "Hmm, That's the first step is like, "Hmm, what do I care about here?" what do I care about here?" what do I care about here?" If you move that earlier, we've aligned If you move that earlier, we've aligned If you move that earlier, we've aligned on that, the code review becomes much on that, the code review becomes much on that, the code review becomes much simpler. simpler. simpler. Number two, moving in too many Number two, moving in too many Number two, moving in too many directions. Again, we're aligning early directions. Again, we're aligning early directions. Again, we're aligning early as a team and individually, I'm like as a team and individually, I'm like as a team and individually, I'm like understanding what I'm working on across understanding what I'm working on across understanding what I'm working on across many agents. If I'm working on a large many agents. If I'm working on a large many agents. If I'm working on a large thing, thing, thing, I've I've understood it initially. So, I've I've understood it initially. So, I've I've understood it initially. So, as a team, we're sharing these plans and as a team, we're sharing these plans and as a team, we're sharing these plans and we're aligning. This is what we're we're aligning. This is what we're we're aligning. This is what we're trying to build and it's easy before trying to build and it's easy before trying to build and it's easy before someone has spent someone has spent someone has spent even a day in AI going deep on some idea even a day in AI going deep on some idea even a day in AI going deep on some idea building a prototype, we can talk about building a prototype, we can talk about building a prototype, we can talk about it early and make sure we're aligned on it early and make sure we're aligned on it early and make sure we're aligned on where we actually taking the system.

  15. where we actually taking the system. where we actually taking the system. Um declaring agent bankruptcy is just Um declaring agent bankruptcy is just Um declaring agent bankruptcy is just not a thing because you've made your not a thing because you've made your not a thing because you've made your agent stateless. So, agent stateless. So, agent stateless. So, the result of their work is in the the result of their work is in the the result of their work is in the stock. If you need to rebuild your human stock. If you need to rebuild your human stock. If you need to rebuild your human context, context, context, you just read the doc. Now, you you just read the doc. Now, you you just read the doc. Now, you understand the state of this project and understand the state of this project and understand the state of this project and you can pick up from there. you can pick up from there. you can pick up from there. And the most important one, And the most important one, And the most important one, humans own the decisions. Like, that's humans own the decisions. Like, that's humans own the decisions. Like, that's what we're solving for. Humans need to what we're solving for. Humans need to what we're solving for. Humans need to own the decisions. own the decisions. own the decisions. That's how we retain ownership of our That's how we retain ownership of our That's how we retain ownership of our software and our products software and our products software and our products and make it like a true expression of and make it like a true expression of and make it like a true expression of what we're trying to create in the what we're trying to create in the what we're trying to create in the world. There are also some other benefits. So, There are also some other benefits. So, when you have this state extracted and when you have this state extracted and when you have this state extracted and you're working from a shared context, you're working from a shared context, you're working from a shared context, you get more parallel agents. It's you get more parallel agents. It's you get more parallel agents. It's easier to work with parallel agents. easier to work with parallel agents. easier to work with parallel agents. Uh you get this durable decision log. Uh you get this durable decision log. Uh you get this durable decision log. So, something uh that's really powerful. So, something uh that's really powerful. So, something uh that's really powerful. I think a lot of people are thinking I think a lot of people are thinking I think a lot of people are thinking about how do we capture all the about how do we capture all the about how do we capture all the decisions going into these sessions? decisions going into these sessions? decisions going into these sessions? A really great solution to that is let's A really great solution to that is let's A really great solution to that is let's pull out all the decisions up front and pull out all the decisions up front and pull out all the decisions up front and agree to them and put them in a place agree to them and put them in a place agree to them and put them in a place that's durable so that we don't have to that's durable so that we don't have to that's durable so that we don't have to have like some LLM summarizing it and have like some LLM summarizing it and have like some LLM summarizing it and maybe picking the wrong things later on.

  16. maybe picking the wrong things later on. maybe picking the wrong things later on. We want to bring that forward so we can We want to bring that forward so we can We want to bring that forward so we can save it and refer to it later. save it and refer to it later. save it and refer to it later. Um and then my favorite one um Um and then my favorite one um Um and then my favorite one um is actually more collaboration. is actually more collaboration. is actually more collaboration. Like Like Like this process, we want we're doing more this process, we want we're doing more this process, we want we're doing more work that is creative, work that is creative, work that is creative, which means we should be collaborating which means we should be collaborating which means we should be collaborating more as engineers. I think the future of more as engineers. I think the future of more as engineers. I think the future of engineering is multiplayer. engineering is multiplayer. engineering is multiplayer. It's going to be multi multiplayer by It's going to be multi multiplayer by It's going to be multi multiplayer by default sooner than we think. default sooner than we think. default sooner than we think. Um Um Um and it's more important than ever and it's more important than ever and it's more important than ever because we're in this moment with AI because we're in this moment with AI because we're in this moment with AI where everything's moving faster than where everything's moving faster than where everything's moving faster than ever. There's a pressure like everyone's ever. There's a pressure like everyone's ever. There's a pressure like everyone's company feels existential, small or big. company feels existential, small or big. company feels existential, small or big. You need to deliver. And the way we get You need to deliver. And the way we get You need to deliver. And the way we get through that as humans is by working through that as humans is by working through that as humans is by working together. And we need tools that help us together. And we need tools that help us together. And we need tools that help us do that. So, here are three So, here are three concrete things you can leave this room concrete things you can leave this room concrete things you can leave this room and do right now. and do right now. and do right now. The first one is to think of your work The first one is to think of your work The first one is to think of your work in terms of planning and polish. in terms of planning and polish. in terms of planning and polish. So, recognizing that there are two gears So, recognizing that there are two gears So, recognizing that there are two gears I'm working in. There's no longer this I'm working in. There's no longer this I'm working in. There's no longer this one focus of implementation. There's uh one focus of implementation. There's uh one focus of implementation. There's uh plan and then polish.

  17. plan and then polish. plan and then polish. Um and notice when you're in a single Um and notice when you're in a single Um and notice when you're in a single session and session and session and you're doing both of these in one you're doing both of these in one you're doing both of these in one session. Notice when you drift from the session. Notice when you drift from the session. Notice when you drift from the planning phase into the polish phase and planning phase into the polish phase and planning phase into the polish phase and is your tool serving you for what you're is your tool serving you for what you're is your tool serving you for what you're trying to do at that moment? trying to do at that moment? trying to do at that moment? Uh Uh Uh number two is to start to treat your number two is to start to treat your number two is to start to treat your plan as a portal to the software system. plan as a portal to the software system. plan as a portal to the software system. So, really treating it as this powerful, So, really treating it as this powerful, So, really treating it as this powerful, malleable tool to say like, what matters malleable tool to say like, what matters malleable tool to say like, what matters to to me for what I'm working on right to to me for what I'm working on right to to me for what I'm working on right now? now? now? Um and asking it to show that to you so Um and asking it to show that to you so Um and asking it to show that to you so that you can make the best decisions that you can make the best decisions that you can make the best decisions possible. possible. possible. And number three is share a plan. Like, And number three is share a plan. Like, And number three is share a plan. Like, don't just don't just don't just you know, write the plan, give it to you know, write the plan, give it to you know, write the plan, give it to your agent, and have them implement it. your agent, and have them implement it. your agent, and have them implement it. Give it to someone on your team. This is Give it to someone on your team. This is Give it to someone on your team. This is like, I feel like very unnatural for a like, I feel like very unnatural for a like, I feel like very unnatural for a lot of people. We always we we think we lot of people. We always we we think we lot of people. We always we we think we know what's going on. know what's going on. know what's going on. But, it's always valuable. You have But, it's always valuable. You have But, it's always valuable. You have smart teammates. They have smart teammates. They have smart teammates. They have great context in their heads. You should great context in their heads. You should great context in their heads. You should tap into that. They will give you good tap into that. They will give you good tap into that. They will give you good feedback. feedback. feedback. Uh it's a really valuable thing to do. That's my talk. I'm Matt. Um I'm the CEO That's my talk. I'm Matt. Um I'm the CEO of Ref. If you want to talk about any of of Ref. If you want to talk about any of of Ref. If you want to talk about any of this stuff, this has all my like this stuff, this has all my like this stuff, this has all my like connection information. Ref is a tool connection information. Ref is a tool connection information. Ref is a tool built for the decision layer.

  18. built for the decision layer. built for the decision layer. Uh we work with all of your existing Uh we work with all of your existing Uh we work with all of your existing implementation tools. implementation tools. implementation tools. Um if you want to see a demo, come you Um if you want to see a demo, come you Um if you want to see a demo, come you can come by our booth or just see me out can come by our booth or just see me out can come by our booth or just see me out there. there. there. Uh and I'd love to talk about this Uh and I'd love to talk about this Uh and I'd love to talk about this stuff, so come say hi. stuff, so come say hi. stuff, so come say hi. Thanks.

Summary

The main theme is the challenges and solutions for teams experiencing a 10x speed increase due to individual AI adoption. Key subjects discussed include too many PRs, moving in multiple directions, and "agent bankruptcy." The practical takeaway is a blueprint for how teams can navigate these issues and move forward cohesively and faster.

View original episode ↗