# Elite AI-driven engineering teams double PR output while others stall | LinearB Blog

> LinearB CTO Yishai Beeri breaks down benchmark data revealing how elite AI-driven engineering teams doubled their merged PR output while others stalled. Learn why measuring PR yield rate and cost per PR proves real AI ROI to finance, and discover how automated AI code review prevents review bottlenecks.

_This is a markdown rendering of a live HTML page on linearb.io, generated for AI/LLM consumption — it is not a markdown-only site. To get the full HTML page instead, request this URL with an explicit `Accept: text/html` header (no wildcard, no markdown preference)._


```json
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    {
      "@type": "ListItem",
      "position": 1,
      "name": "Home",
      "item": "https://linearb.io/"
    },
    {
      "@type": "ListItem",
      "position": 2,
      "name": "Blog",
      "item": "https://linearb.io/blog"
    },
    {
      "@type": "ListItem",
      "position": 3,
      "name": "Elite AI-driven engineering teams double PR output while others stall",
      "item": "https://linearb.io/blog/elite-engineering-teams-double-pr-output-ai-benchmarks"
    }
  ]
}
```

[Home](https://linearb.io/)

/

[Blog](https://linearb.io/blog)

/

Elite AI-driven engineering teams double PR output while others stall

# Elite AI-driven engineering teams double PR output while others stall

![Photo of Andrew Zigler](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Headshot3_d7231cbda7?_a=BAVMn6ID0)

By [Andrew Zigler](https://linearb.io/blog/elite-engineering-teams-double-pr-output-ai-benchmarks#andrew-zigler)

|

August 14, 2026

![Blog_Elite_AI_driven_Teams_2400x1256_fd33821a9d](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Elite_AI_driven_Teams_2400x1256_fd33821a9d?_a=BAVMn6ID0)

The AI question has moved past adoption. Nearly every developer now uses these tools in some form, which means the interesting story is no longer who switched them on. It is who translates that usage into shipped code, and who spends heavily while output stays flat.

[Yishai Beeri](https://linearb.io/dev-interrupted/podcast/linearb-yishai-beeri-2026-ai-productivity-gap-pr-yield), CTO at LinearB, has spent recent months reading the data behind that divide, drawn from millions of pull requests across hundreds of engineering organizations. The picture is sharper than the hype cycle would suggest. Some teams have genuinely doubled their throughput. Others look exactly the same as they did a year ago, despite comparable tool access. Understanding what separates the two groups is now the central problem for engineering leaders.

## AI productivity gap between engineering teams keeps widening

The clearest finding from the mid-year data is a divide that is accelerating. "There's a gap that's opening between teams that are getting significant leverage out of AI and the teams that have merely just switched on or maybe not even really adopted it yet, and that gap is widening at an extremely rapid pace," says Dev Interrupted co-host Ben Lloyd Pearson, reviewing [the benchmark data](https://linearb.io/resources/ai-engineering-productivity-gap?utm%5Fsource=Substack&utm%5Fmedium=referral&utm%5Fcampaign=202607-engineering-prod-gap-guid). That gap is now visible not only company to company but across teams and even individual developers inside the same organization.

Beeri frames it as a question of leverage rather than usage. Coding is only a small part of what developers do, and writing code faster does not automatically mean delivering more of it. Sloppy output that needs rework in two weeks erases the velocity it appeared to create. A new bottleneck in review and acceptance caps the gains no matter how fast the code arrives. "We're seeing a very big divide in the kind of productivity gains that different teams and different organizations and even different developers get," he notes. The teams pulling ahead are the ones solving for those bottlenecks across the whole SDLC, not just accelerating the keyboard. And the timing is specific. From last June through the end of the year, throughput stayed largely stable. Something clicked in January, when frontier model improvements and better tooling finally started translating into measurable output, and the leaders began breaking away.

## AI ROI and cost justification now demand real proof

Finance did not politely join these conversations. "Finance is not just in the room. They came in late, they kicked the door in, frantically trying to get on top of what's happening." The early posture was [tokenmaxxing](https://linearb.io/resources/life-beyond-tokenmaxxing?utm%5Fsource=Substack&utm%5Fmedium=referral&utm%5Fcampaign=beyond-tokenmaxxing), everyone experimenting, nobody wanting to miss the boat. Then the earmarked annual budgets got blown through in two or three months, and the mandate came back from the management offsite.

That mandate is not to cut spend. It is to show value. AI has grown large enough to register on the CFO's radar, moving from a side experiment to five, ten, sometimes twenty percent of dev spend. Layer in recent price hikes as models expose more of the real inference cost, and unlimited experimentation is no longer defensible. "I need to show actual ROI. This is now CFO, CEO-level visibility," Beeri says. Developer sentiment is not enough. Everyone reporting that the tools are great does not answer the finance question. And the rework problem lurks underneath all of it. Tracking token spend alone hides how much refactoring, fixing, and cleaning up of AI-generated messes is happening, and it can quietly erase the exact ROI leaders are being asked to prove.

## PR merge rate reveals true engineering productivity gains

The metric that best captures delivered value is deceptively simple. "How many PRs can I get merged, and normalize that by the number of developers," Beeri says. [PRs merged per developer](https://linearb.helpdocs.io/article/79yuphprpb-merged-frequency-metric) per week is a simplification, like every metric, but it captures whether AI is producing code that actually reaches the codebase rather than just code that gets written.

The data attached to that metric is the headline. Developers in the 90th percentile of AI usage for code that lands in pull requests "year over year, they've more than doubled their output in terms of code that has been merged into production," notes Lloyd Pearson, while developers who do not use AI for their PR code are flat. Today looks like a year ago for them. The uplift correlates down the cohorts, with the top tier seeing two to two-and-a-half times their previous throughput. Notably, the cycles are not much shorter. The gain comes from running more work in parallel. There is an elegant property here too. The obvious ways to game the metric, splitting work and shrinking PRs, are precisely the practices that make reviews easier and code cleaner. Cheating this number means writing better software, so the incentive points in the right direction.

## AI code review workflows quietly boost PR yield rates

Yield rate exposes where AI-generated volume goes to die. Some incoming PRs merge, others are rejected, ignored, or abandoned, and the ratio tells the story. [Human-authored work yields high, often in the high 80s to low 90s. When a human uses AI to write code but stays in charge, yield drops only slightly, two or three points, attributable to larger PRs and the difficulty of reviewing them. Fully autonomous agentic flows collapse to around 30 percent.](https://linearb.io/resources/engineering-productivity-gap) An agent that pulls a ticket, implements it, and pushes a PR with no human owner produces three PRs where only one survives, because nobody is chasing approvals, addressing comments, and driving the change to merge. Getting code shipped is an act of ownership, and detaching the human removes it.

The lever that pushes yield back up is putting AI into the review step, not just the writing step. Basic bugs, style issues, and even hard-to-find defects can be caught before a human ever opens the PR. "Reviewing code, at least a large part of it, can be done by AI," and running that pass first means the human reviewer sees a clean change rather than sloppy work. That reduces cognitive load, protects senior engineers from burning out on an endless queue of messy AI PRs, and lets them apply real expertise where it matters. The measured effect is consistent: "Having AI code review as part of my process, our data shows it ups the yield rate across almost all kinds of PRs by at least two or three percentage points." 

## AI token spend demands new budget management discipline

Companies are setting generous per-developer monthly caps to prevent expensive mistakes rather than to ration daily work. "I've heard of companies using a $2,000 limit or a $1,000 limit per month," and most developers never reach it. The cap is a backstop, not a leash.

The metric maturing on top of those guardrails is cost per PR, which blends human and AI cost into a single figure. A thousand PRs at a hundred thousand dollars is a hundred dollars per PR, and that number falls when throughput rises or bottlenecks like low yield rates get removed. It gives managers a natural way to compare groups, spot where a trend is heading the wrong way, and decide where to intervene. From there the budget behaves like a portfolio. Teams proving high leverage earn more tokens and often more people, because every dollar returns many. Teams still blocked get modest budgets and a directive to learn from the ones who cracked it, fix their quality and review bottlenecks, and prove returns before the spend grows. Coding may not even be their real constraint. And the strategic point is that this line item is not going away. "If it's five or ten percent today, it's going to be twenty or thirty percent tomorrow," Beeri says. Token spend is becoming a growing rival to headcount, and even a recruiting perk, since strong engineers will not join a shop that starves them of budget.

The old model of engineering finance was simple, because headcount was the only cost that mattered. Get good talent and let them work. That world is gone. Managing a dev organization now means balancing people against tokens, steering spend toward proven leverage, and treating output, yield, and cost per PR as one connected system. The teams that win the next year will not be the ones that adopted AI earliest. They will be the ones that turned adoption into leverage, [unblocked the parts of the SDLC](https://linearb.io/resources/measuring-efficiency-in-sdlc) that have nothing to do with typing, and could prove the returns when finance asked.

To hear the full discussion between Yishai Beeri and Ben Lloyd Pearson on engineering ROI, cost per PR, and AI review workflows, listen to the complete episode on the Dev Interrupted podcast.

## Improve developer productivity with LinearB

Find us on

[](https://www.linkedin.com/company/linearb)
[](https://devinterrupted.substack.com/)

![Headshot3_d7231cbda7](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Headshot3_d7231cbda7?_a=BAVMn6ID0)

## Andrew Zigler

Andrew Zigler is a GTM Engineer at LinearB and the host of Dev Interrupted, a twice-weekly podcast and newsletter where 40k+ builders decode the transition to AI-native development and agentic orchestration. A classicist by training with a degree from The University of Texas at Austin, Andrew spent his early career teaching in Japan before channeling his interdisciplinary instincts into the tech world. His polymath background informs everything he builds, from automated workflows to the stories he tells about the seismic shifts reshaping software creation.

### Connect with

[](https://www.linkedin.com/in/andrewzigler)
[](https://substack.com/@zigler)
[](https://x.com/andrewzigler)

## Your next read

[![Cover image for Asana's agentic work management closes the AI productivity gap](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Post_Name_2400x1256_6_e0379266e2?_a=BAVMn6ID0)](https://linearb.io/blog/asana-arnab-bose-agentic-work-management-productivity-gap)

AI

[Asana's agentic work management closes the AI productivity gap](https://linearb.io/blog/asana-arnab-bose-agentic-work-management-productivity-gap)

Asana Chief Product Officer Arnab Bose breaks down how agentic work management closes the widening AI productivity gap between individuals and organizations....

[![Cover image for Your software factory needs a context layer](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/unnamed_4_b35455ae7a?_a=BAVMn6ID0)](https://linearb.io/blog/software-factory-2026-ai-benchmarks-code-review-roi)

AI

[Your software factory needs a context layer](https://linearb.io/blog/software-factory-2026-ai-benchmarks-code-review-roi)

Data from 2.7 million pull requests across 253 engineering organizations shows a widening gap between developers using AI deeply and everyone else. An adoption...

[![Cover image for You can now measure AI ROI with LinearB](https://assets.linearb.io/image/upload/c_limit,w_3840/f_auto/q_auto/v1/Blog_Post_Name_2400x1256_1_dd1f47bae7?_a=BAVMn6ID0)](https://linearb.io/blog/ai-roi-dashboard)

AI

[You can now measure AI ROI with LinearB](https://linearb.io/blog/ai-roi-dashboard)

Today, we're launching the AI ROI dashboard in LinearB, live now for every customer with metrics builder enabled.

## Structured data

_Machine-readable metadata (JSON-LD) embedded in the page for search/AI context — not content rendered on the page itself._

```json
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "LinearB",
  "url": "https://linearb.io/",
  "logo": "https://assets.linearb.io/image/upload/v1715628027/logo-mark-lg.svg",
  "description": "LinearB is the engineering productivity platform that helps engineering leaders prove AI is improving throughput without sacrificing delivery confidence, flow efficiency, or developer experience.",
  "sameAs": [
    "https://www.linkedin.com/company/linearb"
  ],
  "award": [
    {
      "@type": "Award",
      "name": "LinearB is a Leader in the 2026 Gartner® Magic Quadrant™ for Developer Productivity Insight Platforms",
      "dateAwarded": "2026",
      "awardedBy": {
        "@type": "Organization",
        "name": "Gartner®"
      }
    },
    {
      "@type": "Award",
      "name": "Great Place to Work Certification",
      "dateAwarded": "2025-2027",
      "awardedBy": {
        "@type": "Organization",
        "name": "Great Place to Work"
      }
    },
    {
      "@type": "Award",
      "name": "America's Best Startup Employers 2025",
      "dateAwarded": "2025",
      "awardedBy": {
        "@type": "Organization",
        "name": "Forbes Magazine"
      }
    }
  ],
  "hasCertification": [
    {
      "@type": "Certification",
      "name": "SOC 1 Type 2"
    },
    {
      "@type": "Certification",
      "name": "SOC 2 Type 2"
    },
    {
      "@type": "Certification",
      "name": "GDPR Compliance certification"
    },
    {
      "@type": "Certification",
      "name": "ISO 27001"
    }
  ]
}
```

```json
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Elite AI-driven engineering teams double PR output while others stall",
  "url": "https://linearb.io/blog/elite-engineering-teams-double-pr-output-ai-benchmarks",
  "author": {
    "@type": "Person",
    "name": "Andrew Zigler"
  },
  "datePublished": "2026-08-14T07:00:00.000Z",
  "dateModified": "2026-08-14T07:00:00.000Z",
  "image": "https://assets.linearb.io/image/upload/v1787033684/Blog_Elite_AI_driven_Teams_2400x1256_fd33821a9d.png",
  "publisher": {
    "@type": "Organization",
    "name": "LinearB",
    "logo": "https://assets.linearb.io/image/upload/v1777485755/linearb-logo-2026.png"
  },
  "description": "LinearB CTO Yishai Beeri breaks down benchmark data revealing how elite AI-driven engineering teams doubled their merged PR output while others stalled. Learn why measuring PR yield rate and cost per PR proves real AI ROI to finance, and discover how automated AI code review prevents review bottlenecks."
}
```

## More on linearb.io

### Top navigation

- [Book a Demo](https://linearb.io/book-a-demo)
- [AI Code Reviews — Catch security risks, bugs, and spec mismatches](https://linearb.io/platform/ai-code-reviews)
- [AI & Productivity Insights — See how AI tools affect cycle time and delivery speed](https://linearb.io/platform/ai-developer-productivity-insights)
- [Measure AI Impact — Track AI adoption and tie it to delivery outcomes](https://linearb.io/use-case/measure-ai-impact)
- [MCP Server — Chat with your data to spot patterns and boost output](https://linearb.io/platform/mcp-server)
- [Resource Allocation — Cost initiatives and shape your investment strategy](https://linearb.io/platform/resource-allocation)
- [Cost Capitalization — Capitalize engineering costs with audit-ready reports](https://linearb.io/platform/cost-capitalization)
- [Dev Team Management — Set targets and tie throughput to business outcomes](https://linearb.io/platform/goals-and-reporting)
- [DevOps Workflow Automation — Policy-based PR routing, approvals, and tests](https://linearb.io/platform/ai-workflow-governance)
- [AI Powered Support — Unify AI and human code delivery in one clear view](https://linearb.io/use-case/ai-powered-support)
- [Optimization — Surface friction with feedback and MCP insights](https://linearb.io/platform/developer-experience)
- [Reporting — Spot what's working and what needs attention](https://linearb.io/use-case/measuring-developer-experience)
- [Surveys — Turn developer feedback into actionable signals](https://linearb.io/platform/developer-surveys)
- [Platform overview](https://linearb.io/platform/overview)
- [Register now](https://linearb.io/event/the-great-software-factory-debate-2)
- [Customers](https://linearb.io/customers)
- [Pricing](https://linearb.io/pricing)
- [Why choose LinearB — Explore your data. Measure performance. Act to improve it.](https://linearb.io/why-linearb)
- [APEX framework — The operating model for AI-era engineering teams](https://linearb.io/resources/apex-framework)
- [Anti-FAQ — The questions other vendors won't answer](https://linearb.io/why-linearb/anti-faq)
- [Security — Enterprise-grade compliance and zero code access](https://linearb.io/security)
- [Build vs. buy — The hidden cost of building it yourself](https://linearb.io/resources/build-vs-buy)
- [Dev Interrupted Podcast — Conversations with engineering leaders](https://linearb.io/dev-interrupted/podcasts)
- [Reports & Guides — Deep dives on productivity and delivery](https://linearb.io/resources)
- [Webinars — Expert sessions on productivity and AI](https://linearb.io/resources?category=workshops)
- [Metrics Benchmarks — See how your engineering org stacks up](https://linearb.io/resources/software-engineering-benchmarks-report)
- [Blog — Product updates and practical insights](https://linearb.io/blog)
- [Help Center — Documentation, setup, and support](https://linearb.helpdocs.io)
- [API Docs](https://docs.linearb.io/api-overview)
- [Status](https://www.linearbstatus.com/)
- [Integrations](https://linearb.io/integrations)
- [LinearB Library](https://linearb.io/library)
- [Engineering metrics](https://linearb.io/library/engineering-metrics)
- [Platform engineering](https://linearb.io/library/platform-engineering)
- [Engineering glossary](https://linearb.io/library/engineering-glossary)
- [Developer productivity](https://linearb.io/library/developer-productivity)
- [AI in software development](https://linearb.io/library/ai-in-software-development)
- [Engineering management](https://linearb.io/library/engineering-management)
- [Developer experience](https://linearb.io/library/developer-experience)
- [DevOps](https://linearb.io/library/devops)
- [Engineering operations and the context layer](https://linearb.io/library/engineering-operations)
- [Engineering efficiency](https://linearb.io/library/engineering-efficiency)
- [Software delivery](https://linearb.io/library/software-delivery)
- [Research and data](https://linearb.io/library/engineering-benchmarks-and-research)
- [LinearB is a Leader in the 2026 Gartner® Magic Quadrant™ for Developer Productivity Insight Platforms](https://linearb.io/resources/gartner-magic-quadrant-dpi-platforms-2026)
- [Sign in](https://app.linearb.io/login)
- [Enterprise](https://linearb.io/solutions/enterprise)
- [Contact](https://linearb.io/contact-us)
- [About us](https://linearb.io/about-us)
- [Careers](https://linearb.io/careers)
- [Service agreement](https://linearb.io/services-agreement)
- [Privacy policy](https://linearb.io/privacy-policy)
- [DPA](https://linearb.io/data-processing-agreement)
- [Security FAQ](https://linearb.io/security-faq)
- [Substack](https://devinterrupted.substack.com/)

### Footer

_Additional links from the site footer, not repeated from the top navigation above._

- [GitHub](https://github.com/linear-b)
- [LinkedIn](https://www.linkedin.com/company/linearb)
- [Twitter](https://twitter.com/LinearB_Inc)