Over 88% of developers use AI regularly, but AI-assisted pull requests merge at less than half the rate of human-authored code. In this episode, Dan Lines and Ben Lloyd Pearson break down the findings from LinearB's 2026 Engineering Benchmarks Report to reveal how AI is fundamentally reshaping software delivery. They explore the stark behavioral differences between unassisted, AI-assisted, and fully agentic pull requests, highlighting how AI accelerates code generation but exposes massive bottlenecks in the review process. Tune in to learn why organizations must prioritize AI readiness, data quality, and context engineering before they can translate raw AI adoption into actual business impact.
Show Notes
- LinearB 2026 Engineering Benchmarks Report: Download the full 48-page report to see where your metrics stand.
Transcript
(Disclaimer: may contain unintentionally confusing, inaccurate and/or amusing transcription errors)
[00:00:00] Dan Lines: Hey, what's up everyone? Welcome back to Dev Interrupted. I'm your host, Dan Lines, LinearB, COO, and co-founder. Today's episode will focus on the 2026 engineering benchmark from LinearB. This year's report, it's our biggest, our most comprehensive, yet it's an amazing, amazing report. We're going to take a look at what the data says about engineering teams and really specifically how AI is fundamentally reshaping the way we build software.
[00:00:35] Dan Lines: And to help me walk through all this data, I have my fellow co-host and LinearB's Director of AI Innovation, Ben Lloyd Pearson, BLP, to answer all of my questions about this year's report. Ben, always great to share an episode with you. Welcome to the show.
[00:00:57] Ben Lloyd Pearson: Yeah. Thank, thanks for having [00:01:00] me. Uh, it's always a lot of fun to just share all the cool research we're doing at LinearB, so a lot of cool data for us to unpack today.
[00:01:07] Dan Lines: Yeah. Amazing. And you and I were talking before the show. You did, you came on, Dev Interrupted to do this I think like two years ago. Maybe even
[00:01:17] Ben Lloyd Pearson: TH
[00:01:18] Dan Lines: years
[00:01:18] Ben Lloyd Pearson: three years ago.
[00:01:19] Dan Lines: three years ago, and so I think it will be pretty interesting to see how this has progressed. Obviously a lot of it has to do with ai, but how do you feel about now coming on three years later and doing the same thing, but with an updated report?
[00:01:36] Ben Lloyd Pearson: Oh man, it, I mean, it's wild to think about just how different the world is today versus when I, when I first started here at LinearB, I mean, AI was barely even something we were talking about at that point. And now it's like all we can talk about and I feel like everything I'm doing is now being impacted by it.
[00:01:52] Ben Lloyd Pearson: So, you know, of course. So is this report and there's gonna be a lot of new AI stuff that we're gonna talk about today, which is pretty awesome.
[00:01:59] Dan Lines: [00:02:00] Yeah, it's really cool. I mean, I, I read through everything in there and definitely like, it's gonna be exciting to get into that AI stuff that's really popping out. A lot of good insights there. But I guess probably the best way, let's start with an overview. start out with an overview summary. maybe you can give us an overview of what makes either this report, uh, unique or maybe some of the things that are kind of like, uh, jumping out at you, uh, for this year.
[00:02:28] Ben Lloyd Pearson: Yeah, well, like you said, it's, it's the most comprehensive analysis we've ever done. Um, 8.1 million pull requests, about 4,800 engineering teams spread across 42 different countries, so pretty large scale of data that we're dealing with here, and what really sets this year apart from. Is that we've introduced a completely new dimension to this report around AI productivity and specifically how to measure it.
[00:02:54] Ben Lloyd Pearson: So for the first time, we're not just looking at all the traditional software delivery metrics, you know, those are still there, we're still [00:03:00] reporting on them, but we're also examining how AI is impacting engineering workflows across the board. So, and on top of that, we've combined, uh, for the first time ever.
[00:03:10] Ben Lloyd Pearson: Qualitative data along with our, or excuse me, qualitative analysis along with our quantitative data, uh, where we surveyed a bunch of engineering leadership or engineering leaders about how they're using AI within their organization. both to understand the data behind what's changing, but then also the perception of, of the leaders that are running these things.
[00:03:31] Ben Lloyd Pearson: And if I had to pick one thing that is sort of the, the top line that everyone should take away from this is that. AI adoption is effectively maximized at this point. It's, it's almost universal. So in, in our findings, 88.3% of developers now use AI regularly. So that's, you know, at least multiple times a week, which is up from 70, just under 72% when we last surveyed this back in early 2024.
[00:03:57] Ben Lloyd Pearson: But I wanna make, I wanna stress one [00:04:00] really critical. Point, and that is that adoption does not equal impact. So AI generated prs are behaving completely different than ones that are unassisted by ai. They're larger, they wait longer for reviews, and they merge at less than half the rate of human authored code.
[00:04:20] Ben Lloyd Pearson: So what do I really want people to take away is that AI is accelerating code generation. I, I think we all know that at this point, but it's also exposing bottlenecks everywhere else in the SDLC, primarily with things like reviews, testing, governance, organizational readiness. There's still a lot that, uh, needs to be solved outside of code generation.
[00:04:41] Dan Lines: Yeah, man. Uh, you mentioned a few things there. One thing that I really, uh, like about the report when, and hopefully like all of our listeners will, will read through this, we do have the qualitative responses now. So as you read through the report, you can see what engineering leaders are saying and commenting on and that type of thing.
[00:04:59] Dan Lines: [00:05:00] So I thought that was really cool. The other thing that I really liked about the report on the AI side is, it doesn't just talk about ai, I guess generically as being like one single unit,
[00:05:13] Ben Lloyd Pearson: Yeah.
[00:05:14] Dan Lines: there's the agentic pull requests, there's the AI assisted, there's the non-AI assisted and the comparison there.
[00:05:24] Ben Lloyd Pearson: Exactly.
[00:05:25] Dan Lines: maybe you can explain, you know, what, what's the difference between like the AI assisted and the fully agentic just, uh, for our listeners.
[00:05:33] Ben Lloyd Pearson: Yeah. Yeah. So we have a lot of data that compares these three classifications of prs. So, um, you know, on the simplest side you have unassisted prs. So this is something where a human authors it, they don't use AI to generate code at all. Probably don't even use it to ideate. They just write the code, sort of as we've always done for many years.
[00:05:52] Ben Lloyd Pearson: Um, and then submit it. Um, one level up from that is AI assisted prs. So this is where the code is, is authored by a [00:06:00] human, but is significantly shaped with AI tools, whether that be the actual generation of the code or, you know, maybe planning for generating the code or, or you know, research that type of stuff.
[00:06:11] Ben Lloyd Pearson: And then at the highest level you have agentic pr. So those are pull requests that are created entirely by. An AI agent, um, definitely the most, um, or the, the less, the least mature of these three categories, but that's agents like Devon, copilot, stuff like that when they, they generate the pull request themselves rather than having a human lead.
[00:06:30] Ben Lloyd Pearson: The effort.
[00:06:31] Dan Lines: Yeah. And we're gonna get into, uh, some of those D differences. 'cause I
[00:06:35] Ben Lloyd Pearson: Yep.
[00:06:36] Dan Lines: really cool how the report broke that out. Um, and obviously that that is new to the report. What else is new compared to previous, uh, years for the report?
[00:06:47] Ben Lloyd Pearson: Yeah, so the biggest thing is probably the new section that we introduced CO, that focuses exclusively on AI productivity insights, and we, when we've broken this down into a few parts, so AI code in the SDLC, so this is [00:07:00] how AI generated in AI assisted prs behave compared to manual or unassisted pull requests.
[00:07:07] Ben Lloyd Pearson: Uh, the DevEx of ai. So this is how developers are experiencing AI adoption and their level of trust with it. Um, and then the last big addition related to AI is, uh, a new survey that that, uh, gauges the state of AI readiness among respondents. So this is a look at whether organizations have the foundations to make AI successful.
[00:07:28] Ben Lloyd Pearson: And if you've read this year's DORA report. This will, this will be very familiar with you because we tried to build upon the re the great research that they're doing over there. Um, related to all this new ai, AI stuff. We've also added a new benchmark this year for acceptance rate, and that's not the typical one of, of how much.
[00:07:46] Ben Lloyd Pearson: How many lines of code did you accept or how, what percentage of, of submit of recommendations from an a, uh, coding assistant did you accept? It's not that. Um, instead we're measuring the percentage of pull requests that get merged [00:08:00] within 30 days. So this is really relevant for AI generated code where.
[00:08:05] Ben Lloyd Pearson: Things like ownership and acceptance patterns can be completely different than your typical pull request. And, uh, you know, just just to preview before we get into it too much, the data on this is really striking. You know, manual PRS merge at about 80, 84 point half percent, so about 84% of all the
[00:08:24] Ben Lloyd Pearson: unassisted prs that are open get merged across all of our users. Um, but AI prs merge at just 32.7%, which is less than half of that rate. So a pretty stark difference between those two numbers.
[00:08:40] Dan Lines: Yeah, and it's like super cool that we have that data. That's some of some of the ones that popped out to me. And yeah, when we say accept acceptance rate, like usually, uh, that's something that's pushed by the vendors like copilot and all of them saying how much, you know, code that we suggested to developers that they actually [00:09:00] accept.
[00:09:00] Dan Lines: I think this is even actually more useful. It's more so saying of all this code that's being generated. How much of it is actually making it into customer's hands or like into production? And there's such a big difference between the agent, the assisted, you know, fully developer created. Uh, I know we're gonna dive into it, but I really wanted to highlight that one as well.
[00:09:22] Ben Lloyd Pearson: Yeah. And, and I mean, when you're thinking about lines of code that is generated by an ai, if you measure that as your acceptance rate, it, it doesn't actually mean that any of that code ultimately gets accepted. The developer might just be generating it and then deleting half of it because they don't like it.
[00:09:34] Ben Lloyd Pearson: You know? Um, this is a great, our version of this metric is a great way to understand. Of all the things that AI is generating for your organization, which of them are actually useful enough to be deployed to production? Like that's a far more meaningful metric to track.
[00:09:48] Dan Lines: Yeah, high value metric. okay. I know there's. A few other things that are different or maybe different in how we're analyzing, uh, the data [00:10:00] we already talked about. Okay. Now there's three types of prs,
[00:10:04] Ben Lloyd Pearson: Mm-hmm.
[00:10:04] Dan Lines: agentic AI prs, there's the AI assisted prs, there's the unassisted. So like fully developer prs. Um, I think there's something about an AI readiness matrix Uh, yeah, maybe you talk to us
[00:10:19] Ben Lloyd Pearson: Yeah.
[00:10:19] Dan Lines: and if there's anything else that you wanna dive into.
[00:10:22] Ben Lloyd Pearson: Yeah, I'll, I'll, I'll cover the readiness matrix, but I do wanna point out one, just, just one really interesting factor about those three classifications of PRS that we described. So again, that's unassisted, that is AI assisted and agentic ai prs. Um, you know, like, like I said, we wanted to attract differences in behaviors between all of these and, and some examples of what we were able to.
[00:10:44] Ben Lloyd Pearson: To discover, is that for, for example, with AI assisted prs, they tend to be about two and a half times larger than unassisted prs when you're looking at the, the P 75. So the 75th percentile. Um, but what's more striking [00:11:00] is that AI prs have a pickup time that is more. Five times longer than unassisted prs.
[00:11:07] Ben Lloyd Pearson: So they're basically waiting idle for review. Much, much longer than manually generated code. So again, very stark differences in what we're seeing here.
[00:11:17] Ben Lloyd Pearson: . Um, and then on top of that, uh, we've also introduced this AI readiness matrix, again, based on Dora research, um, within Dora, within the latest Dora report. They have, you know, these seven categories of success criteria with ai. You know, a big, a big story they're pushing with Dora is how AI amplifies both the good and the bad.
[00:11:36] Ben Lloyd Pearson: So if you have, for example, bad version control practices, when you start using ai, it's gonna amplify all those bad sides of it. So, um, we surveyed, a bunch of engineering leaders, to understand whether the typical organization has the foundational capabilities. So there's things like data quality, version control, maturity, clear AI policies, the [00:12:00] things that actually takes to make AI successful.
[00:12:03] Ben Lloyd Pearson: Um, and there's some pretty eye-opening findings around that too. Like, uh, for example, 65% of companies lack dependable data quality. and on top of that AI policy alignment is sharply polarized. So, um, you know, there, there is a very big difference between organizations that are strong at this stuff and, and not so strong.
[00:12:23] Dan Lines: Okay. amazing stuff. New report. A lot of new things going on in the world with ai. The report captures all of the differences. The first one that I'd like to dive into is what you were giving us, I think, a little appetizer on with. The unassisted versus the assisted versus fully agentic. Maybe you can recap again the data behind that.
[00:12:48] Dan Lines: But I, what I wanna get into is, let's figure out why uh, 'cause the behaviors are like so different there. So maybe
[00:12:55] Ben Lloyd Pearson: Yeah.
[00:12:56] Dan Lines: of the data and then let's talk about like why that's happening.
[00:12:59] Ben Lloyd Pearson: Yeah. So to [00:13:00] recap, you know, AI assisted prs are being accepted and merged at about half the rate of manually generated prs, and we had a couple of other data points that that sort of indicate why this might be happening. So again, looking at the P 75 level, the the 75th percentile, the typical AI, assisted PR has about four, has a little over 400 lines of code, which is pretty large.
[00:13:24] Ben Lloyd Pearson: It's not gigantic, but it is large. and that's compared to 157 lines of code for unassisted prs. Um, and then age agentic AI PRs are kind of in the middle at about 290 ish. Lines of code. So, um, when you think about it, you know, our, one of our recommendations we've always had at LinearB is that, um, you wanna try to keep your PRs below 300 lines of code because that's sort of a, a chunk of, of information that is relatively, it's, it's manageable for a human to keep all that information in our head and make.
[00:13:54] Ben Lloyd Pearson: Clear decisions about it and review it promptly and efficiently. Um, so when you have pull [00:14:00] requests that are getting, you know, 400 plus lines of code, it, it increases a lot of the mental tax on people who have to review the code. Um, you know, so people are reviewing more, there's more files they have to look at.
[00:14:12] Ben Lloyd Pearson: There's more parts of the system that get affected when PRs get bigger and just an overall higher complexity. Um, and one of the, and I I really love the qualitative survey we introduced with this report 'cause it does add a lot of flavor to all of the data points that we've been gathering to help us understand why these sorts of things are happening so we can, and, and the surveyor questions around this kept coming back to very similar issues, like AI tends to make changes that are larger than the scope of what was requested from them.
[00:14:43] Ben Lloyd Pearson: They're very verbose at times. Um, they often just do more than a human would, uh, which is, I mean, when you think about it, creating more code is effectively free for an ai. It takes a lot of mental power for a human to do it, so it's just naturally much easier for an AI to just generate larger [00:15:00] volumes.
[00:15:00] Dan Lines: Yeah, I mean, I, I, I think the takeaway is like, okay, if I am a developer, I'm not using ai. My PR sizes are generally smaller. Probably because, you know, it's pretty intensive to actually code. You gotta write, think about it, write out all the code. As a developer, you're usually trying to get like the most done possible with the least amount of code. I mean, that's really hard. Hands on work for ai. These prs, in one sense could be getting bloated and the downstream impact of that. I'm thinking about quality. I'm thinking about quality both from like, I don't know, maybe like bugs and security and that kind of stuff. But I'm also interested. In, uh, the reviewer portion, meaning if a human's gonna go and have to review that, now maybe AI's also doing the review, but if a developer's gonna review it, well my review's definitely not gonna be as good as if it's smaller. then you also mentioned something about maybe some scope creep in there? Like is it [00:16:00] doing more than it's really asked to do? Um, it feels like it's putting so much pressure, like on the review process. Is that how you feel about it? Or like whether you, whether you, what are your thoughts?
[00:16:11] Ben Lloyd Pearson: Yeah, I mean that, that is the bottleneck right now that I think most organizations that are adopting AI are struggling with the most right now. So, you know, I already mentioned that, that these, that AI generated prs wait over five times longer. It's, it is actually about 5.25 times longer to be picked up for a review.
[00:16:31] Ben Lloyd Pearson: Um, so that's a representation of a little over a thousand minutes, which a smarter person than me would've to figure out how many hours that is. Uh, for AI generated prs versus about 200 minutes for an unassisted PR to get picked up for review. But here's where things get really kind of weird. Um, but it, it will make sense.
[00:16:48] Ben Lloyd Pearson: I think if we break this down,
[00:16:49] Dan Lines: Okay.
[00:16:50] Ben Lloyd Pearson: once someone starts reviewing. An AI generated code, they, it tends to get reviewed much faster. Um, in fact, it's, [00:17:00] it's, uh, the typical AI assisted PR gets reviewed in about 194 minutes compared to 252 minutes for manual pr. So while they take longer to get picked up, they get reviewed faster.
[00:17:13] Dan Lines: Oh, okay. Well, yeah, we gotta break that, that's really interesting. the pickup time is longer, meaning, okay, let's say I see that there's a PR out there and it needs to be reviewed. If it's assisted by ai, it takes longer for that review to actually begin. But once the review begins, it's actually faster.
[00:17:35] Ben Lloyd Pearson: Yep. Do you have a guess why?
[00:17:38] Dan Lines: Well, I mean, I, I, I would say that I'll, I'll start with the part that makes, uh, most sense to me. 'Cause right now we are talking about assisted versus unassisted, but there's that third category of full of fully agentic, right?
[00:17:55] Ben Lloyd Pearson: Mm-hmm.
[00:17:56] Dan Lines: That third category of fully agentic. I could see [00:18:00] why those wouldn't be picked up as fast because it's like. Who, who owns that?
[00:18:07] Ben Lloyd Pearson: Yeah. This isn't my code. This isn't my friend's code.
[00:18:11] Dan Lines: BLP if you opened up a PR that you worked hard on and then it was like a sign to me. And we're buddies, and I know you're trying to get this done. I have a lot of incentive to go help out my boy. Like to, okay, I gotta get BLP's,
[00:18:25] Ben Lloyd Pearson: Yep.
[00:18:26] Dan Lines: code out there. That makes sense to me.
[00:18:27] Dan Lines: 'cause there's like a human component in ownership. Now if ai, which I have no like, uh, relation to opens up to me, what's a random pr like, I'm not really incentivized to pick it up like that makes sense to me.
[00:18:41] Ben Lloyd Pearson: Yep.
[00:18:42] Dan Lines: I'm the only thing that I could say. Now, let's say BLP, you're using, uh, assistance in your, so let's say that you're using copilot, but it would still be under your name right when the PR goes up.
[00:18:58] Ben Lloyd Pearson: Right.
[00:18:58] Dan Lines: incentivized [00:19:00] to go get it, we also said that the PR is larger. And if the PR is larger in general, me as a reviewer, I'm gonna be less incentivized to wanna start that. That's the only thing I can think of. Is it that or is there other stuff behind the
[00:19:16] Ben Lloyd Pearson: you, you're definitely on to, to something. Uh, and I think there's a little more to this too. So, and again, this was really where it was really great for us to lean on qualitative data this time as well, because one thing that came up in multiple survey respondents was that, um, you know, a lot of people are just hesitant to open AI prs.
[00:19:33] Ben Lloyd Pearson: You know, they're uncertain about them. They're, they don't know what the mental load is gonna be on that. They, they, they have concerns about trust with the ai, like, is it gonna contain errors or missing context? And then, and then it becomes my problem. Like, if I look at it and I start reviewing it, then, then it becomes my problem.
[00:19:48] Ben Lloyd Pearson: You know, so I think that explains part of why they take so long to get picked up for review. Um, but then once, once the reviewer does pick up, uh, you know, they tend to just scan for high level issues rather than [00:20:00] like, deeply understanding the change. You know, one of the respondents to our survey said that, you know, a larger amount of our code is slipping into production without
[00:20:09] Ben Lloyd Pearson: proper review. So rubber stamping, like that's what is starting to happen more and more. Um, and then we also heard concerns about how, you know, AI often gives up mistaken suggestions and non-working code. And, and I've been hearing a lot, particularly in, in the last year about the struggles with trust when it comes to ai, uh, performance.
[00:20:29] Ben Lloyd Pearson: You know, do you actually trust that what it's doing is the right thing and that it has good security and good quality and, and all of those factors? So. I really think it is, it is a combination of bigger pull requests coming in from ar a ai, the lack of ownership over who, who that PR belongs to. Um, coupled with, uh, you know, just this sense that, you know, it's very challenging to, to, to modify, to work with AI after it's already done a thing. So, you know, it can sometimes be frustrating if you have [00:21:00] to repeatedly tell AI that it got something wrong and, and push back and have it, you know, continually correct itself. So I think it's a, it's a, it is a sort of confluence of many factors that are contributing to, uh, to this interesting dichotomy between pickup and review time.
[00:21:14] Dan Lines: Yeah, I think that's one of the most interesting things in the report. The reason, uh, that I think it is interesting is if we take a step back, adopting AI only matters for an engineering, organization because the promise is getting productivity right, on the other side. Like if you're a CTO, VP, director you are adopting these AI tools with the intent that you're releasing more features, releasing more features on time, providing real value. Right. a lot of this stuff with okay, agentic prs are being opened, but I think I, I don't have the exact data, less than 50% actually even make it out to production. Right. And, [00:22:00] and, uh, or maybe, you know, there's some quality issues on the other side. I think like one of the reasons to look at the report. Is to understand for yourself, okay, if I have an AI strategy, are the hiccups that are being seen in the industry right now about actually getting, uh, that AI code out into production?
[00:22:19] Dan Lines: Then therefore, what can I do about it? I think that's like one of the coolest things about this whole report.
[00:22:23] Ben Lloyd Pearson: Yep.
[00:22:24] Dan Lines: I also had an idea to run by you, BLP,
[00:22:29] Ben Lloyd Pearson: Oh.
[00:22:29] Dan Lines: and I don't think, I don't think it's in the report, it might be. we said something was like for the assisted, the assisted prs, right? The AI assisted prs.
[00:22:45] Dan Lines: They're being, they're bigger, but they're being reviewed faster. Is that true?
[00:22:49] Ben Lloyd Pearson: Yes.
[00:22:51] Dan Lines: Is there data overlaid on that? That also says, are those ones being AI reviewed well? [00:23:00] And the reason that I'm, I'm bringing that up is if you're in a situation that, okay, let's say that we're using co-pilot as an organization or we're using Cursor as an organization and I have the same tool, creating code and reviewing the code, I have a theory that it's not, the review is not gonna find as many issues. And therefore, let's say I am a human developer and I'm going to do the review, and I'm really reliant on that AI review. might say, Hey, it didn't find too many issues. And therefore I think it's good. I'm just gonna do a cursory overview versus if it's all human created code and then I have an AI run on top of it in the review, and then the reviewers there, they might see some stuff like, whoa, it found a lot of things and the review's gonna take longer. I don't know that it's actually in the report, [00:24:00] but I have like a theory. A little bit here about, okay, if co-pilot's doing both the co-generation and the review, it's not gonna find too much more in the review and the review's gonna actually go faster.
[00:24:11] Ben Lloyd Pearson: Yeah, that's, that's interesting. We, we don't, we don't have data specifically for this and maybe it would be a good addition for next year, particularly when you consider just how common AI code reviews are becoming. Um, you know, I've been saying for a little while now that the things like AI code reviews, AI pull request descriptions, those are like the very first, like ubiquitous use cases for AI within the SDLC.
[00:24:33] Ben Lloyd Pearson: Um, even, you know, even beyond like code generation, I think if you're thinking about the benefits you can gain from ai, those code reviews are, are really where a lot of the benefits stack up. I I think what it really comes down to is the level of trust. Like do you have trust that the AI code review.
[00:24:49] Ben Lloyd Pearson: You have on your repos is actually catching things that your team should be taking action on. Um, and if the answer is, is no, then you're probably gonna have just as much scrutiny as you did [00:25:00] before, and you're probably just gonna ignore whatever the AI code review, uh, is telling you. And if the answer is yes, you do trust it, then when it tells you everything's green, you're just gonna say, well, everything's probably green then.
[00:25:12] Ben Lloyd Pearson: And if it warns you, that stuff is orange or red. Then you're gonna, you're gonna accept that and try to fix it. So, and, and I think this really just points out, you know, how, um, your organizational practices are really the thing that defines success with ai. So if you aren't able to tell your, your AI code review or your AI assistance, um, what your organization's standards.
[00:25:35] Ben Lloyd Pearson: The things that you expect all prs to do, um, even down to like the specific components within your tech stack, like how to build with those and how to use them and, and your security requirements for all of those things. Um, it, it, it, it just comes down to having the configuration granularity to achieve a level of trust that would allow, um, developers to, to either believe that the code ai, code reviews are [00:26:00] good or, um, be, uh, yeah, effectively that.
[00:26:03] Dan Lines: Yeah, these are the rules and policies that, uh, make your strategy actually work. Like first you have to have a strategy. have to have an AI strategy that allows your workflow, uh, from code creation out to production actually be smooth. If you don't have a strategy, you'll probably be the same or worse.
[00:26:24] Dan Lines: And then, yeah, all those rules or automations that you're putting in place like that matters so much. I wanna keep on the AI topic 'cause it's probably most interesting, but change, uh, up a little bit. What does the report say about type of code created? Meaning, um, is AI creating a bunch of new code or is it more like doing like bug fixes and that kind of stuff?
[00:26:47] Dan Lines: Like do we have any data on that?
[00:26:48] Ben Lloyd Pearson: Yeah, this, this, uh, this data couldn't be any more clear than the state of that actually. Um, AI is almost exclusively generating new code. [00:27:00] So when you look at the, the P 75 of unassisted PRs, they have a refactor rate of about 0.37. So that means about 37% of the code in the PR is, is a refactor of existing code.
[00:27:14] Ben Lloyd Pearson: Um, but with AI assisted prs, it's almost zero. It's barely. It's, it's practically negligible, um, which means that AI is creating a ton of new code paths rather than focusing on improving legacy code. Um, and it's potentially not helping solve any problems associated with technical debt. It may actually be creating more tech debt than it is solving for.
[00:27:38] Ben Lloyd Pearson: Um, I, I think that this is something that may get solved over time as the technology behind this improves. Like you just start training the model to be, to be more critical about generating new code and to focus on leveraging existing code more. It's probably also something to say about the context that you feed into these models.
[00:27:56] Ben Lloyd Pearson: You know, if, if you just kind of give it free reins and, and code [00:28:00] generation is a piece of cake for ai, then it's just gonna keep creating new code because to it, to, to an LLM, it doesn't matter how much code you have. So, yeah, it's, this is one of the areas where it's, you know, you every organization should understand at this point that AI is probably generating more new code than your organization has ever seen before.
[00:28:19] Dan Lines: Yeah, that's really, really interesting. Let's see if we can dive in there more. do you think it's because, okay, so AI is definitely generating new code as compared to like cleaning up tech debt.
[00:28:36] Ben Lloyd Pearson: Yeah.
[00:28:37] Dan Lines: that's a takeaway. Okay. Do you think it's because that's what it's being instructed to do, like the rules and the prompts and that type of thing? Is it because that's what it's better at? Like it's better at creating new codes versus looking at existing code and modifying it? do you think it's like, or is it something else? Like, I'm trying to just [00:29:00] like, uh, or is that how I don't know us humans as developers are saying like, Hey, what I need the most help with is actually creating a bunch of new code.
[00:29:08] Dan Lines: Like I can go and refactor stuff on my own. I'm trying, trying to
[00:29:11] Ben Lloyd Pearson: Yeah,
[00:29:12] Dan Lines: it.
[00:29:12] Ben Lloyd Pearson: I, I think it's the, the, the opposite of your first idea. So it, it's not being instructed to create new code. It's, it's instead not being instructed to focus on using let or existing code. Right? So this is where if you just take AI and throw it at your code base it doesn't care what exists today unless you tell.
[00:29:31] Ben Lloyd Pearson: That it needs to care about that. Um, and I think that's really one of the challenges that most engineering organizations will need to solve to really see the big productivity gains, like this type of challenge, like that granular configurability of, if you're gonna use this library, you have to access it this way.
[00:29:47] Ben Lloyd Pearson: Or if you're gonna do this type of function, you must use this library we've already created for it over here. Um, so that way if it encounters issues trying to implement that stuff, it's doing it within the, the framework of your [00:30:00] organization. And instead of just saying, well, I'm just gonna do it all new, because that's the easiest and the quickest way to solve the problem.
[00:30:06] Dan Lines: getting
[00:30:07] Ben Lloyd Pearson: Um,
[00:30:08] Dan Lines: like
[00:30:08] Ben Lloyd Pearson: exactly.
[00:30:09] Dan Lines: ai, okay, here's all the context. Here's uh, all these diff uh, here's, here's our policy of how we work at this company.
[00:30:18] Ben Lloyd Pearson: Yep.
[00:30:18] Dan Lines: we use these libraries. So make sure you. You're saying it's more like feeding all of that in is what would be needed. Most companies are probably not doing that, or we're maybe not as good at it yet. Therefore, it's, it likes doing. Okay. You're telling me to do like greenfield stuff, like build from scratch type of thing?
[00:30:37] Ben Lloyd Pearson: Yep. Yeah. If, if, if, if prompt engineering with a phrase of the last year, I think context engineering is the phrase of the next year. So, um, instead of focusing on better prompting and, and all of that, you should be focused on the context that you're building and supplying to these models because context can be, can be applied universally, it can be applied granularly.[00:31:00]
[00:31:00] Ben Lloyd Pearson: Um, it's, it's really the thing that is going to determine success when you're adopting ai. And, and this actually gets in, if I can just go straight into my next point on AI readiness. 'cause this me. Perfectly into what I wanted to talk to you
[00:31:11] Dan Lines: for
[00:31:12] Ben Lloyd Pearson: on that. So, you know, we, we, we survey as a part of our survey, we, um, you know, we borrowed from the DORA research where they have their seven categories of AI readiness.
[00:31:21] Ben Lloyd Pearson: We condensed it slightly down to six just 'cause we already had a, a pretty large survey. Um, and we, we asked. These engineering leaders that responded on a, on a series of, uh, uh, challenges that every organization faces. So things like version control, practices, delivery habits, internal and pla, internal tooling and platform reliability.
[00:31:44] Ben Lloyd Pearson: Um, these are all things that you need to be good at to be successful with ai. Um, and we just asked, you know, how. robust are these systems for you? So for example, we asked, um, do you have clear and communicated AI policies? And for that [00:32:00] particular one, uh, we had very polarizing responses.
[00:32:03] Ben Lloyd Pearson: um. so about 60% of respondents believe that they have a clear ai, at least somewhat clear AI policy, while about 26% believe the opposite. And then about 14% were somewhere in the middle. So this is sort of like one of the first steps that you really should have on your AI readiness journey, so to speak. You know, just telling your developers like, what tools are approved, how should you use them? How do you request new tools? Where are we applying AI within our SDLC? Those types of things. And then the other one, getting back to our narrative of context that we keep talking about. Um, a majority of organizations indicate that their data is not ready for AI models.
[00:32:44] Ben Lloyd Pearson: About 65% of organizations say they have data, data problems, and, and I think this is honestly probably the biggest challenge of 2026 for even just putting engineering aside. I think anyone who is trying to adopt ai, whether it's for software [00:33:00] engineering or marketing or sales or customer success, whatever it may be.
[00:33:04] Ben Lloyd Pearson: If you don't have high quality data that provides the context that the AI needs for whatever sort of problem you're trying to solve, you're gonna, you're going to encounter significant issues. This is where hallucination becomes a problem. This is where bad results become a problem, and this is the type of thing that erodes trust in ai.
[00:33:22] Ben Lloyd Pearson: So, so clear and communicated AI policy and data hygiene and availability. Those are the two big challenges that I think, uh, the industry as a whole is facing right now.
[00:33:32] Dan Lines: This report is so freaking cool. I guess it's like, you know, all based on this AI revolution that everyone's in, but like I love how it has the data and then also the qualitative responses from the leaders and stuff like that was like really hooking me while I was reading through. So now, you know, we talked a lot of numbers and that type of stuff on the pod so far. As we're moving into, and maybe [00:34:00] we're in like planning, maybe we're in a planning mode. I'm an engineering leader. I know I gotta get this AI stuff working for me. Um, not just doing stuff, but actually getting to an output. What are your thoughts on, uh, the things I should be focusing on or any like, uh, things that I might trip me up as an engineering leader?
[00:34:22] Dan Lines: What should I be thinking about?
[00:34:24] Ben Lloyd Pearson: Yeah. So first of all, don't confuse adoption with impact. Just because everyone is using AI today, it doesn't mean that you're getting real value from it yet. Uh, you know, except as, as I mentioned, 88% and other studies have said somewhere like around 90 to 95% of developers are using AI every week at this point.
[00:34:44] Ben Lloyd Pearson: Um, adoption is basically universal, as I I mentioned. but impact productivity gains, they aren't yet. so you need to measure things like, like acceptance rates, you know, our version of it where you're measuring the amount of AI code that's actually making it into production. Um, you need to measure [00:35:00] your review patterns, like where the bottlenecks showing up, what, what are causing, what things are causing the biggest bottlenecks, uh, within your teams, and then what are your actual delivery outcomes like, are you delivering.
[00:35:11] Ben Lloyd Pearson: More features to your, to the company, to the, to your customers or whatever, whatever sort of business outcomes your, your executives are concerned about, is AI actually supporting, uh, those needs rather than just focusing on, you know, raw usage statistics. And then the other point I'll make is that, you know, you need to fix your foundations before scaling ai.
[00:35:32] Ben Lloyd Pearson: Uh, looking at this AI readiness, and we're gonna have a lot more content around this concept of AI readiness, AI enablement over the coming months because we think it's a really important topic right now. But you need to fix all of those foundational problems like high quality data, clear policies, reliable internal tooling.
[00:35:50] Ben Lloyd Pearson: If you have all of those things, AI will amplify the benefits of them. And if you have problems with them, AI will amplify the problems that they create. So it [00:36:00] will make the worst worse, and it will make the best better. Uh, so those, those are the two things that I, I think every engineering or leader out there should be focused on, uh, for the next year.
[00:36:09] Dan Lines: Yeah, I, that all makes sense. And I, I guess what I, I would add on, like, my takeaway is I also think 2026 is gonna be a year, uh, around quality. I mean by that is, like you said, if AI is amplifying everything. It's also creating bigger prs. It's also maybe doing some scope creep and creating things that we don't even need from the story and that type of stuff. what we're focusing on at LinearB, and what I would encourage all the listeners, is to think about quality. So for example, how many security issues are now being created by ai? How many bugs, how many maintainability issues
[00:36:50] Ben Lloyd Pearson: Yep.
[00:36:51] Dan Lines: our tech debt, uh, getting bigger or smaller? And then the last thing that I'd re-emphasize that you said, BLP, is, [00:37:00] yes, the impact is super, super, important, also can I start giving AI more context so that it's doing the right thing more often, which then would reduce the quality problem. So that would be my concern or the thing to, to watch out for. Just because it's doing more doesn't mean it's doing better.
[00:37:22] Ben Lloyd Pearson: Exactly.
[00:37:25] Dan Lines: All right, cool. Is there anything else that we need to hit on before we go to the out outro here?
[00:37:30] Ben Lloyd Pearson: Uh, no, I just, just wanna remind our listeners go read the report. It's, it's 48 pages packed with tons of data that we, a lot of which we weren't, aren't even able to scratch the surface on in this short podcast episode. Um, we love producing this. It's, we always learn a ton ourselves as we do it, and I know our audience does as well.
[00:37:49] Ben Lloyd Pearson: So, yeah, go download the report, check it out. You'll learn a lot.
[00:37:52] Dan Lines: All right. Well, thanks so much, Ben. Yeah. This, this re this year's report. It's a total, uh, game changer. Yeah. [00:38:00] It's not just about benchmarking performance anymore. It's also about all the stuff we talked about, understanding how AI is fundamentally reshaping software delivery. If you or your team want to see where your metrics stand and understand how AI is impacting your organization, be sure to check out the 2026 Engineering Benchmarks report at LinearB dot io. put a link in the show notes. Uh, thanks again, Ben, and thanks everyone for listening. We'll see you next week.
