How to Generate Mergeable Code with a Context Engine — Peter Werry, Unblocked

What we do at Unblocks is we build a What we do at Unblocks is we build a context engine. I just want to do a context engine. I just want to do a context engine. I just want to do a quick sound check at the back to make quick sound check at the back to make quick sound check at the back to make sure everyone can hear me fine. Can you sure everyone can hear me fine. Can you sure everyone can hear me fine. Can you guys Yeah, we're good. Awesome. So at a guys Yeah, we're good. Awesome. So at a guys Yeah, we're good. Awesome. So at a high level, a context engine delivers high level, a context engine delivers high level, a context engine delivers organizational context to both your organizational context to both your organizational context to both your human workers and now increasingly your human workers and now increasingly your human workers and now increasingly your agents. Okay. So why why is that agents. Okay. So why why is that agents. Okay. So why why is that important? Before we go too deep on the important? Before we go too deep on the important? Before we go too deep on the mechanics of how a context engine works, mechanics of how a context engine works, mechanics of how a context engine works, I just want to talk briefly about the I just want to talk briefly about the I just want to talk briefly about the problem. problem. problem. So, what we're going to do is we're So, what we're going to do is we're So, what we're going to do is we're going to hop into our time machines and going to hop into our time machines and going to hop into our time machines and we're going to travel back to the before we're going to travel back to the before we're going to travel back to the before times uh before agents and uh discuss a times uh before agents and uh discuss a times uh before agents and uh discuss a little bit about what we used to do as little bit about what we used to do as little bit about what we used to do as humans uh before agents came into the humans uh before agents came into the humans uh before agents came into the picture. picture. picture. And so for years um you were the context And so for years um you were the context And so for years um you were the context layer. You um had to go and do things layer. You um had to go and do things layer. You um had to go and do things like this. you had to find things you like this. you had to find things you like this. you had to find things you were looking for, trolled all over were looking for, trolled all over were looking for, trolled all over different data sources, different different data sources, different different data sources, different discussions taking place. Um, and then discussions taking place. Um, and then discussions taking place. Um, and then through the codebase of course to try to through the codebase of course to try to through the codebase of course to try to build up tribal knowledge and, uh, build up tribal knowledge and, uh, build up tribal knowledge and, uh, throughout time as your code base throughout time as your code base throughout time as your code base progressed, um, you'd be, you know, progressed, um, you'd be, you know, progressed, um, you'd be, you know, fighting incidents and things like that.

fighting incidents and things like that. fighting incidents and things like that. And your organization over time builds And your organization over time builds And your organization over time builds up battle scars um from all these all up battle scars um from all these all up battle scars um from all these all these different things building code uh these different things building code uh these different things building code uh documenting architecture and and dealing documenting architecture and and dealing documenting architecture and and dealing with outages and things like that. with outages and things like that. with outages and things like that. But now um we have a new problem because But now um we have a new problem because But now um we have a new problem because uh as we introduce agents to the picture uh as we introduce agents to the picture uh as we introduce agents to the picture um they suffer from all of these um they suffer from all of these um they suffer from all of these challenges except for one thing. Agents challenges except for one thing. Agents challenges except for one thing. Agents are like new employees. are like new employees. are like new employees. they reset their knowledge every time they reset their knowledge every time they reset their knowledge every time you start a new task. Okay? And so you you start a new task. Okay? And so you you start a new task. Okay? And so you can think of an agent like an expert can think of an agent like an expert can think of an agent like an expert software engineer um who's a new software engineer um who's a new software engineer um who's a new employee on boarding for the first time. employee on boarding for the first time. employee on boarding for the first time. Every time they have to rediscover your Every time they have to rediscover your Every time they have to rediscover your code base, how your organization builds code base, how your organization builds code base, how your organization builds tests um and how they deploy software tests um and how they deploy software tests um and how they deploy software with each and every task. with each and every task. with each and every task. Uh can I just uh put a put a show of Uh can I just uh put a put a show of Uh can I just uh put a put a show of hands for everyone that's seen this hands for everyone that's seen this hands for everyone that's seen this slide before by Vim?

slide before by Vim? slide before by Vim? So this is kind of like u this is a good So this is kind of like u this is a good So this is kind of like u this is a good way to view where people are on what we way to view where people are on what we way to view where people are on what we call like the AI maturity curve. Um call like the AI maturity curve. Um call like the AI maturity curve. Um starting at the the far left uh this is starting at the the far left uh this is starting at the the far left uh this is kind of representative of autocomplete kind of representative of autocomplete kind of representative of autocomplete back in the GBT35 days. You know back in the GBT35 days. You know back in the GBT35 days. You know remember co-pilot and things like that. remember co-pilot and things like that. remember co-pilot and things like that. Um and then you know kind of move on to Um and then you know kind of move on to Um and then you know kind of move on to using cursor. Um and then from there using cursor. Um and then from there using cursor. Um and then from there you're you're think you're talking about you're you're think you're talking about you're you're think you're talking about how you can start to solve the context how you can start to solve the context how you can start to solve the context problem. So some people are building problem. So some people are building problem. So some people are building organizational wikis. Just smile if if organizational wikis. Just smile if if organizational wikis. Just smile if if this is kind of um bringing up memories this is kind of um bringing up memories this is kind of um bringing up memories for you. Um and then you know all these for you. Um and then you know all these for you. Um and then you know all these things are great except that uh how do things are great except that uh how do things are great except that uh how do you give agents access to this and what you give agents access to this and what you give agents access to this and what are the compounding problems that the are the compounding problems that the are the compounding problems that the scaling problems as you move forward scaling problems as you move forward scaling problems as you move forward well if you give MCP and skills to your well if you give MCP and skills to your well if you give MCP and skills to your agents agents agents um to teach them how to navigate and um to teach them how to navigate and um to teach them how to navigate and build context and that's kind of where build context and that's kind of where build context and that's kind of where uh people are today most people they're uh people are today most people they're uh people are today most people they're at the sort of stage four to five level at the sort of stage four to five level at the sort of stage four to five level okay okay okay and uh they understand that context is and uh they understand that context is and uh they understand that context is the bottleneck and they're trying to the bottleneck and they're trying to the bottleneck and they're trying to build solutions to solve it for their build solutions to solve it for their build solutions to solve it for their engineering teams. So looking ahead uh engineering teams. So looking ahead uh engineering teams. So looking ahead uh to all the way to eight with software to all the way to eight with software to all the way to eight with software factories. This is kind of where the factories. This is kind of where the factories. This is kind of where the puck is going. I'm not sure if if folks puck is going. I'm not sure if if folks puck is going. I'm not sure if if folks were at the keynote this morning, but um were at the keynote this morning, but um were at the keynote this morning, but um it's it's all about like delivery of it's it's all about like delivery of it's it's all about like delivery of context and unknown and unknowns. And context and unknown and unknowns. And context and unknown and unknowns. And this becomes increasingly important as

this becomes increasingly important as this becomes increasingly important as people start thinking about full people start thinking about full people start thinking about full automation of agents. they just can't automation of agents. they just can't automation of agents. they just can't operate without organizational context. operate without organizational context. operate without organizational context. They get lost. So, you know, like that's the real So, you know, like that's the real problem. Access to information doesn't problem. Access to information doesn't problem. Access to information doesn't equal understanding. Um I I know that equal understanding. Um I I know that equal understanding. Um I I know that folks are probably familiar with folks are probably familiar with folks are probably familiar with claude.md claude.md claude.md um and and uh and wiki layouts and all um and and uh and wiki layouts and all um and and uh and wiki layouts and all these things. If you attach a wiki, it these things. If you attach a wiki, it these things. If you attach a wiki, it still doesn't tell the agent where the still doesn't tell the agent where the still doesn't tell the agent where the information is that it needs. It can information is that it needs. It can information is that it needs. It can search for things in the wiki, but then search for things in the wiki, but then search for things in the wiki, but then what happens is it'll suffer from what happens is it'll suffer from what happens is it'll suffer from something that uh radiologists something that uh radiologists something that uh radiologists uh call satisfaction of search. So, this uh call satisfaction of search. So, this uh call satisfaction of search. So, this is a term in radiology is a term in radiology is a term in radiology where you look at an X-ray and you're where you look at an X-ray and you're where you look at an X-ray and you're trying to find a region um that might be trying to find a region um that might be trying to find a region um that might be an indicator for cancer. Okay? And you an indicator for cancer. Okay? And you an indicator for cancer. Okay? And you discover like one discover like one discover like one indicator and if you stop there uh you indicator and if you stop there uh you indicator and if you stop there uh you might miss other important indicators might miss other important indicators might miss other important indicators that might you know lead to diagnosis of that might you know lead to diagnosis of that might you know lead to diagnosis of even more uh issues. So this is what even more uh issues. So this is what even more uh issues. So this is what happens with agents. They don't they happens with agents. They don't they happens with agents. They don't they they find something that they they think they find something that they they think they find something that they they think is correct and then they stop. Um the is correct and then they stop. Um the is correct and then they stop. Um the the other thing about agents is that the other thing about agents is that the other thing about agents is that they don't distill understanding.

they don't distill understanding. they don't distill understanding. They can look around, they can find They can look around, they can find They can look around, they can find information, but they they don't information, but they they don't information, but they they don't understand how all the pieces fit understand how all the pieces fit understand how all the pieces fit together because without doing that leg together because without doing that leg together because without doing that leg work ahead of time. Um, they don't work ahead of time. Um, they don't work ahead of time. Um, they don't understand how, you know, your understand how, you know, your understand how, you know, your dependencies interact with each other dependencies interact with each other dependencies interact with each other and how your architecture and sort of and how your architecture and sort of and how your architecture and sort of future planning is going to scope the future planning is going to scope the future planning is going to scope the work that it does next. And so some some work that it does next. And so some some work that it does next. And so some some people will then ask, well, what if we people will then ask, well, what if we people will then ask, well, what if we just take the entire codebase and all of just take the entire codebase and all of just take the entire codebase and all of our architecture documents and just slam our architecture documents and just slam our architecture documents and just slam it into the context window. Um, and then it into the context window. Um, and then it into the context window. Um, and then yes, maybe like your agents will reason yes, maybe like your agents will reason yes, maybe like your agents will reason about everything all at once. And in about everything all at once. And in about everything all at once. And in practice, that that of course doesn't practice, that that of course doesn't practice, that that of course doesn't work. Um, not just because you've got work. Um, not just because you've got work. Um, not just because you've got way more organizational context than can way more organizational context than can way more organizational context than can fit into a context window, even one fit into a context window, even one fit into a context window, even one that's a million tokens in size. Um, but that's a million tokens in size. Um, but that's a million tokens in size. Um, but it it it causes the agent to get it it it causes the agent to get it it it causes the agent to get distracted. When you're working on a distracted. When you're working on a distracted. When you're working on a task, you want task specific flow. Um, task, you want task specific flow. Um, task, you want task specific flow. Um, and so your agents will get distracted and so your agents will get distracted and so your agents will get distracted easily if you give them things that easily if you give them things that easily if you give them things that cause them to look this way in that way. cause them to look this way in that way. cause them to look this way in that way. Um, and it'll just waste tokens and Um, and it'll just waste tokens and Um, and it'll just waste tokens and time. So, in this morning's keynote, um, time. So, in this morning's keynote, um, time. So, in this morning's keynote, um, Tariq from Claude Code mentioned unknown Tariq from Claude Code mentioned unknown Tariq from Claude Code mentioned unknown unknowns. I just want to uh harp on that unknowns. I just want to uh harp on that unknowns. I just want to uh harp on that phrase again. And it can be phrased a phrase again. And it can be phrased a phrase again. And it can be phrased a different way, which is finding the different way, which is finding the different way, which is finding the things that really matter.

things that really matter. things that really matter. And so this is what your agent can see And so this is what your agent can see And so this is what your agent can see at the top of the iceberg. They can see at the top of the iceberg. They can see at the top of the iceberg. They can see the code and they can operate on the the code and they can operate on the the code and they can operate on the code. What they don't see are things code. What they don't see are things code. What they don't see are things like the actual intent, the team like the actual intent, the team like the actual intent, the team conventions, past decisions, things that conventions, past decisions, things that conventions, past decisions, things that you've discussed in Slack, for example, you've discussed in Slack, for example, you've discussed in Slack, for example, uh architecture rationale, and so on. uh architecture rationale, and so on. uh architecture rationale, and so on. And that's why your agents need a And that's why your agents need a And that's why your agents need a context engine to get real work done. context engine to get real work done. context engine to get real work done. So, I'm going to now uh attempt a live So, I'm going to now uh attempt a live So, I'm going to now uh attempt a live demo. And hopefully the demo gods are demo. And hopefully the demo gods are demo. And hopefully the demo gods are kind. Um, so I want to pop back up kind. Um, so I want to pop back up kind. Um, so I want to pop back up conceptually. Oops, I think I'm on the conceptually. Oops, I think I'm on the conceptually. Oops, I think I'm on the wrong tab. We'll get to that one in a wrong tab. We'll get to that one in a wrong tab. We'll get to that one in a sec. sec. sec. So for now, So for now, So for now, sorry about that. And here we are. So I'm going to ask a question as if I'm So I'm going to ask a question as if I'm a, you know, I'm a human and I want to a, you know, I'm a human and I want to a, you know, I'm a human and I want to get some information about my codebase.

get some information about my codebase. get some information about my codebase. And, you know, the human layer hasn't And, you know, the human layer hasn't And, you know, the human layer hasn't gone away. We talk about agents and gone away. We talk about agents and gone away. We talk about agents and their need for context, but um humans their need for context, but um humans their need for context, but um humans are still asking questions about the are still asking questions about the are still asking questions about the codebase and we need that level of codebase and we need that level of codebase and we need that level of understanding because ultimately the understanding because ultimately the understanding because ultimately the accountability stops with us. When you accountability stops with us. When you accountability stops with us. When you hit merge on a PR, you need to hit merge on a PR, you need to hit merge on a PR, you need to understand what it's doing um and you understand what it's doing um and you understand what it's doing um and you need to understand how the architecture need to understand how the architecture need to understand how the architecture works. So this question I asked here um works. So this question I asked here um works. So this question I asked here um is about an internal component of our is about an internal component of our is about an internal component of our system called the source mark engine and system called the source mark engine and system called the source mark engine and you can see that it uh is able to you can see that it uh is able to you can see that it uh is able to articulate it fairly well. um articulate it fairly well. um articulate it fairly well. um understands the architecture. This this understands the architecture. This this understands the architecture. This this diagram here is uh is generated. So it diagram here is uh is generated. So it diagram here is uh is generated. So it this diagram doesn't exist. Um it just this diagram doesn't exist. Um it just this diagram doesn't exist. Um it just figures it out based on the um the way figures it out based on the um the way figures it out based on the um the way the code operates today and then some the code operates today and then some the code operates today and then some proposals for future architecture. proposals for future architecture. proposals for future architecture. And then uh what's really important is And then uh what's really important is And then uh what's really important is that you show your work. This is a trust that you show your work. This is a trust that you show your work. This is a trust building thing more than anything, but building thing more than anything, but building thing more than anything, but it allows people to see if um if the it allows people to see if um if the it allows people to see if um if the answer is maybe not entirely correct, answer is maybe not entirely correct, answer is maybe not entirely correct, then you can in look into the uh the then you can in look into the uh the then you can in look into the uh the knowledge base that you have and make knowledge base that you have and make knowledge base that you have and make corrections.

corrections. corrections. Increasingly agents are doing this for Increasingly agents are doing this for Increasingly agents are doing this for you. you. you. So now um what I want to show you is So now um what I want to show you is So now um what I want to show you is another place where humans spend their another place where humans spend their another place where humans spend their time which is in Slack and this is where time which is in Slack and this is where time which is in Slack and this is where a lot of the decisions get made of a lot of the decisions get made of a lot of the decisions get made of course. course. course. So I can do something like this. So I can do something like this. So I can do something like this. And uh unblocked will sit and kind of And uh unblocked will sit and kind of And uh unblocked will sit and kind of listen for things that are things that listen for things that are things that listen for things that are things that can chime in on when it provides a high can chime in on when it provides a high can chime in on when it provides a high degree of Oh, sorry. We went to the degree of Oh, sorry. We went to the degree of Oh, sorry. We went to the wrong You guys can't see that. Thank wrong You guys can't see that. Thank wrong You guys can't see that. Thank you, Claire. you, Claire. you, Claire. Oh, come on down. Let's see if I can Oh, come on down. Let's see if I can Oh, come on down. Let's see if I can bring it up. There we go. bring it up. There we go. bring it up. There we go. Perfect. So I can ask questions like Perfect. So I can ask questions like Perfect. So I can ask questions like this in unblocked and if it thinks it this in unblocked and if it thinks it this in unblocked and if it thinks it can chime in on the answer then it will can chime in on the answer then it will can chime in on the answer then it will chime in. Otherwise I can just chime in. Otherwise I can just chime in. Otherwise I can just um address unblocked directly and ask um address unblocked directly and ask um address unblocked directly and ask the same question and when it thinks that it has an answer and when it thinks that it has an answer to give then it will give an answer and to give then it will give an answer and to give then it will give an answer and so we can get um so we can get um so we can get um quite a bit of interesting content there quite a bit of interesting content there quite a bit of interesting content there from unblocked. Thank you. Unblocked.

I'm going to switch up I'm going to switch up and show you the the really interesting and show you the the really interesting and show you the the really interesting thing which is the agents. Okay. So, um thing which is the agents. Okay. So, um thing which is the agents. Okay. So, um in in that question, the source mark in in that question, the source mark in in that question, the source mark engine, I'm not sure if people picked engine, I'm not sure if people picked engine, I'm not sure if people picked up, but there was a little thing at the up, but there was a little thing at the up, but there was a little thing at the bottom there that said, you know, bottom there that said, you know, bottom there that said, you know, there's some optimization opportunities. there's some optimization opportunities. there's some optimization opportunities. Um so what I did here is I went into Um so what I did here is I went into Um so what I did here is I went into claw code and I asked it um without claw code and I asked it um without claw code and I asked it um without using unblocked to um using unblocked to um using unblocked to um uh generate uh a plan to optimize the uh generate uh a plan to optimize the uh generate uh a plan to optimize the source mark calculator and it did that source mark calculator and it did that source mark calculator and it did that and it happily went and you know and it happily went and you know and it happily went and you know searched through the code and and tried searched through the code and and tried searched through the code and and tried to figure out how the algorithm works to figure out how the algorithm works to figure out how the algorithm works and so on. Um and it it reached a and so on. Um and it it reached a and so on. Um and it it reached a conclusion that's great you know it does conclusion that's great you know it does conclusion that's great you know it does a pretty good job um but you know it a pretty good job um but you know it a pretty good job um but you know it maybe could do a little bit better. So, maybe could do a little bit better. So, maybe could do a little bit better. So, I asked that question again uh using I asked that question again uh using I asked that question again uh using unblock this time and it it really kind unblock this time and it it really kind unblock this time and it it really kind of nails the the nuances because it of nails the the nuances because it of nails the the nuances because it picks up on the the uh PRs that we um picks up on the the uh PRs that we um picks up on the the uh PRs that we um where we discussed future possibilities where we discussed future possibilities where we discussed future possibilities for improvement. um some Slack for improvement. um some Slack for improvement. um some Slack conversations that we had and uh of conversations that we had and uh of conversations that we had and uh of course you know notion and architecture course you know notion and architecture course you know notion and architecture documents and it shows its work and this documents and it shows its work and this documents and it shows its work and this is really important because um all of is really important because um all of is really important because um all of these things here the sources come back these things here the sources come back these things here the sources come back to Claude and then Claude knows exactly to Claude and then Claude knows exactly to Claude and then Claude knows exactly where to jump to next if it needs to where to jump to next if it needs to where to jump to next if it needs to elaborate on that context. And so I just elaborate on that context. And so I just elaborate on that context. And so I just want to show you what the impact of that want to show you what the impact of that want to show you what the impact of that is. So if I um Whoops.

is. So if I um Whoops. is. So if I um Whoops. Thank you. If I pull up usage here, you Thank you. If I pull up usage here, you Thank you. If I pull up usage here, you can see that with unblocked, uh, the can see that with unblocked, uh, the can see that with unblocked, uh, the total cost was, you know, subd dollar to total cost was, you know, subd dollar to total cost was, you know, subd dollar to create the plan. Uh, took about a create the plan. Uh, took about a create the plan. Uh, took about a minute. Ignore the wall clock time minute. Ignore the wall clock time minute. Ignore the wall clock time because I've had this open for about an because I've had this open for about an because I've had this open for about an hour. But, um, it's about a minute. And hour. But, um, it's about a minute. And hour. But, um, it's about a minute. And then if I look at um the usage without then if I look at um the usage without then if I look at um the usage without unblocked, you can see that it's about 2 unblocked, you can see that it's about 2 unblocked, you can see that it's about 2 minutes. And and and it costs more to minutes. And and and it costs more to minutes. And and and it costs more to generate all that context. Now, the generate all that context. Now, the generate all that context. Now, the reason that happens is because it has to reason that happens is because it has to reason that happens is because it has to do more work. It has to look around. has do more work. It has to look around. has do more work. It has to look around. has to discover things. Um, and this to discover things. Um, and this to discover things. Um, and this compounds, not only does it have to do compounds, not only does it have to do compounds, not only does it have to do more work to discover things, it doesn't more work to discover things, it doesn't more work to discover things, it doesn't discover the right things. So, when you discover the right things. So, when you discover the right things. So, when you get further down in your execution, it get further down in your execution, it get further down in your execution, it may be operating on the wrong plan or may be operating on the wrong plan or may be operating on the wrong plan or the wrong assumptions. And then you have the wrong assumptions. And then you have the wrong assumptions. And then you have to go back and you have to loop over and to go back and you have to loop over and to go back and you have to loop over and over again. So, the real value of a over again. So, the real value of a over again. So, the real value of a context engine is not like the upfront context engine is not like the upfront context engine is not like the upfront cost on these short tasks. It's the cost on these short tasks. It's the cost on these short tasks. It's the compounding effect. Um the the other compounding effect. Um the the other compounding effect. Um the the other Tariq from Sonar mentioned this in the Tariq from Sonar mentioned this in the Tariq from Sonar mentioned this in the keynote this morning and it's true like keynote this morning and it's true like keynote this morning and it's true like the loops compound and you have to be the loops compound and you have to be the loops compound and you have to be like um uh efficient the entire way like um uh efficient the entire way like um uh efficient the entire way through with your context. I'm just through with your context. I'm just through with your context. I'm just going to jump back to going to jump back to going to jump back to Safari and I'm going to point out um Safari and I'm going to point out um Safari and I'm going to point out um some really interesting things. So we some really interesting things. So we some really interesting things. So we also have a a code review agent.

also have a a code review agent. also have a a code review agent. And when we say um you know And when we say um you know And when we say um you know organizational context, we're talking organizational context, we're talking organizational context, we're talking about more than just the underlying about more than just the underlying about more than just the underlying data. Uh we're talking about real data. Uh we're talking about real data. Uh we're talking about real intelligence. So what unblock does is it intelligence. So what unblock does is it intelligence. So what unblock does is it looks at um not like it looks at pull looks at um not like it looks at pull looks at um not like it looks at pull request data and there are other data request data and there are other data request data and there are other data sources for this and it generates a sources for this and it generates a sources for this and it generates a series of best practices that help align series of best practices that help align series of best practices that help align agents to your codebase. But we thought agents to your codebase. But we thought agents to your codebase. But we thought that this would be really helpful to that this would be really helpful to that this would be really helpful to surface for the review agent as well. So surface for the review agent as well. So surface for the review agent as well. So what you can see here is um what you can see here is um what you can see here is um it unblock chimed in and then Richie it unblock chimed in and then Richie it unblock chimed in and then Richie here said, "Oh, that's cool. That's here said, "Oh, that's cool. That's here said, "Oh, that's cool. That's something I would say." And that's something I would say." And that's something I would say." And that's because that actually was something he because that actually was something he because that actually was something he said. So it surfaced the uh the previous said. So it surfaced the uh the previous said. So it surfaced the uh the previous comments. Richie's one of the senior comments. Richie's one of the senior comments. Richie's one of the senior engineers and we use the sort of engineers and we use the sort of engineers and we use the sort of seniority or expertise as a signal um to seniority or expertise as a signal um to seniority or expertise as a signal um to boost uh comments that are important. boost uh comments that are important. boost uh comments that are important. Okay. Okay. Okay. So another uh interesting interaction by So another uh interesting interaction by So another uh interesting interaction by Richie, he uh discovered that the number Richie, he uh discovered that the number Richie, he uh discovered that the number of code review issues that were being of code review issues that were being of code review issues that were being surfaced dropped uh precipitously surfaced dropped uh precipitously surfaced dropped uh precipitously and he was debugging it with unblocked.

and he was debugging it with unblocked. and he was debugging it with unblocked. Um he got all the way to the bottom and Um he got all the way to the bottom and Um he got all the way to the bottom and realized what roughly what the problem realized what roughly what the problem realized what roughly what the problem was and then asked unblocked to fix it. was and then asked unblocked to fix it. was and then asked unblocked to fix it. Now this this is something that we have Now this this is something that we have Now this this is something that we have internally um you know that we're internally um you know that we're internally um you know that we're experimenting with. Um, so unblocked uh experimenting with. Um, so unblocked uh experimenting with. Um, so unblocked uh can run as an agent in the cloud. Um, can run as an agent in the cloud. Um, can run as an agent in the cloud. Um, but what's really cool about this is but what's really cool about this is but what's really cool about this is that it has all your organizational that it has all your organizational that it has all your organizational context at its fingertips and the context at its fingertips and the context at its fingertips and the results are are pretty magical. So it results are are pretty magical. So it results are are pretty magical. So it can do things like generate this PR um, can do things like generate this PR um, can do things like generate this PR um, and then what you'll see here is that and then what you'll see here is that and then what you'll see here is that not only does it generate the fix, it not only does it generate the fix, it not only does it generate the fix, it also is able to relate it to the all the also is able to relate it to the all the also is able to relate it to the all the conversations that were happening. So conversations that were happening. So conversations that were happening. So this PR was created because and you read this PR was created because and you read this PR was created because and you read that context thing. It's mind-blowing. that context thing. It's mind-blowing. that context thing. It's mind-blowing. After this PR, we switched to uh Claude After this PR, we switched to uh Claude After this PR, we switched to uh Claude 48 and it dropped a ton in issues 48 and it dropped a ton in issues 48 and it dropped a ton in issues because of the behavior is quite a bit because of the behavior is quite a bit because of the behavior is quite a bit different. So then it said Richie different. So then it said Richie different. So then it said Richie directly correlated the drop. Now what's directly correlated the drop. Now what's directly correlated the drop. Now what's this thing here? Let's click on it. It this thing here? Let's click on it. It this thing here? Let's click on it. It is a Slack conversation. So, it found is a Slack conversation. So, it found is a Slack conversation. So, it found the Slack conversation, correlated all the Slack conversation, correlated all the Slack conversation, correlated all of that, you know, past history back of that, you know, past history back of that, you know, past history back again, and then we ended up with a with again, and then we ended up with a with again, and then we ended up with a with a final PR.

So, um, I'm going to I've got only a few So, um, I'm going to I've got only a few minutes left. I'm just going to close minutes left. I'm just going to close minutes left. I'm just going to close this out really quickly. We have a uh a this out really quickly. We have a uh a this out really quickly. We have a uh a couple of open- source projects that are couple of open- source projects that are couple of open- source projects that are kind of interesting if people want to kind of interesting if people want to kind of interesting if people want to play with them. One is the document play with them. One is the document play with them. One is the document query engine. That was, uh, something query engine. That was, uh, something query engine. That was, uh, something that I talked about on Monday in my that I talked about on Monday in my that I talked about on Monday in my workshop. Um, I may uh talk about it workshop. Um, I may uh talk about it workshop. Um, I may uh talk about it again tomorrow, but I just want to give again tomorrow, but I just want to give again tomorrow, but I just want to give folks a sense of what this thing does. folks a sense of what this thing does. folks a sense of what this thing does. Um, whoops. Um, whoops. Um, whoops. If you want to play with it, it's open If you want to play with it, it's open If you want to play with it, it's open source, so you can just download it and source, so you can just download it and source, so you can just download it and have it go. It basically runs over your have it go. It basically runs over your have it go. It basically runs over your um uh GitHub repository, ingests uh your um uh GitHub repository, ingests uh your um uh GitHub repository, ingests uh your your historical pull requests, and then your historical pull requests, and then your historical pull requests, and then uh synthesizes a schema based on the uh synthesizes a schema based on the uh synthesizes a schema based on the documents that it can sample. Um and documents that it can sample. Um and documents that it can sample. Um and then from there you can issue any kind then from there you can issue any kind then from there you can issue any kind of queries that you like and get all of queries that you like and get all of queries that you like and get all kinds of insights out of it through the kinds of insights out of it through the kinds of insights out of it through the agent chat. You can ask all kinds of agent chat. You can ask all kinds of agent chat. You can ask all kinds of questions. Um and then lastly the questions. Um and then lastly the questions. Um and then lastly the engineering social graph. So this is the engineering social graph. So this is the engineering social graph. So this is the thing that I was talking about earlier thing that I was talking about earlier thing that I was talking about earlier that helps us pin down expertise and that helps us pin down expertise and that helps us pin down expertise and team relationships. Um so what you can team relationships. Um so what you can team relationships. Um so what you can see here is this sort of like the rough see here is this sort of like the rough see here is this sort of like the rough breakdown of our team structure at breakdown of our team structure at breakdown of our team structure at Unblocked. As you can see we're a fairly Unblocked. As you can see we're a fairly Unblocked. As you can see we're a fairly small team. Um and so we've got these um small team. Um and so we've got these um small team. Um and so we've got these um uh these clusters of people and how they uh these clusters of people and how they uh these clusters of people and how they relate to each other indicates the kind relate to each other indicates the kind relate to each other indicates the kind of um review relationships that they of um review relationships that they of um review relationships that they have. So these are you know these lines have. So these are you know these lines have. So these are you know these lines show like we review each other's code.

show like we review each other's code. show like we review each other's code. Um Um Um we can then cluster that and generate we can then cluster that and generate we can then cluster that and generate team labels for that or show the team labels for that or show the team labels for that or show the coverage across your codebase. This is coverage across your codebase. This is coverage across your codebase. This is really cool. you can see kind of where really cool. you can see kind of where really cool. you can see kind of where the holes are, where you might be the holes are, where you might be the holes are, where you might be lacking expert coverage. Um, and that's lacking expert coverage. Um, and that's lacking expert coverage. Um, and that's exactly what we use within the context exactly what we use within the context exactly what we use within the context engine itself. engine itself. engine itself. All right, one last thing we have uh for those that one last thing we have uh for those that want a taste of what a context engine want a taste of what a context engine want a taste of what a context engine can do but don't want to sign up for can do but don't want to sign up for can do but don't want to sign up for unblocked right away. Um you can use uh unblocked right away. Um you can use uh unblocked right away. Um you can use uh something that we call the context something that we call the context something that we call the context engine simulator which will basically engine simulator which will basically engine simulator which will basically build up a context behind the scenes on build up a context behind the scenes on build up a context behind the scenes on a per task basis and then use that a per task basis and then use that a per task basis and then use that context uh to to drive the task. It'll context uh to to drive the task. It'll context uh to to drive the task. It'll do it with context and without context do it with context and without context do it with context and without context so that you can see what the differences so that you can see what the differences so that you can see what the differences might be. might be. might be. This is a QR code for that if you want This is a QR code for that if you want This is a QR code for that if you want to just take a quick snap.

Awesome. And I'll just land Awesome. And I'll just land on a quote from one of our customers. on a quote from one of our customers. on a quote from one of our customers. 50% fewer tokens, faster triage, better 50% fewer tokens, faster triage, better 50% fewer tokens, faster triage, better answers. And that's exactly what a answers. And that's exactly what a answers. And that's exactly what a context engine can do. context engine can do. context engine can do. One last shout out um before we end. My One last shout out um before we end. My One last shout out um before we end. My colleague Brandon is giving a talk in colleague Brandon is giving a talk in colleague Brandon is giving a talk in 10 minutes uh at room 2020. um he's 10 minutes uh at room 2020. um he's 10 minutes uh at room 2020. um he's going to speak to in a lot more detail going to speak to in a lot more detail going to speak to in a lot more detail about some of the higher level things about some of the higher level things about some of the higher level things that context engines can do. I'm going that context engines can do. I'm going that context engines can do. I'm going to run over there right after this and I to run over there right after this and I to run over there right after this and I think all of you should follow me. think all of you should follow me. think all of you should follow me. Awesome. Oh, and don't forget to get a Awesome. Oh, and don't forget to get a Awesome. Oh, and don't forget to get a coconut.