{ "apiVersion": "v1", "topic": { "id": "code-review", "title": "How is AI used in code review?", "summary": "AI that reviews code changes, finds problems before merge, or picks the human reviewer.", "group": "Building software", "aliases": [ "review bot", "pr review", "reviewing code" ] }, "passages": [ { "id": "ramp-93-prs-now-automatically-handled-review-buddy-2026", "text": "93% of our PRs are now automatically handled by Review Buddy so that the highest quality engineers are spending time on the 7% of PRs that truly matter,", "speaker": "Geoff Charles", "role": "Chief Product Officer (CPO)", "organization": "Ramp", "aiVendor": false, "work": { "title": "What product looks like when coding is solved | Geoff Charles (Ramp CPO)", "type": "talk", "date": "2026-09-25", "publisher": "Lenny's Podcast (YouTube channel); Geoff Charles's talk at the Lenny and Friends Summit, recorded 2026-09-10", "locator": "09:34", "url": "https://www.youtube.com/watch?v=ZG8Mf3P9xzI" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [ { "metric": "function-agent-share", "value": 0.93, "unit": "share", "period": "share of PRs, as of 2026-09", "definition": "engineering: pull requests automatically handled by Review Buddy, Ramp's AI code reviewer (unit of work: PRs)", "level": "unclear" } ], "context": null, "url": "https://the-ai-record.com/passages/ramp-93-prs-now-automatically-handled-review-buddy-2026" }, { "id": "uber-ureview-today-analyzes-over-90-weekly-65-2025-2", "text": "uReview today analyzes over 90% of the weekly ~65,000 diffs (equivalent of pull requests) landed at Uber. Engineers who interact with the tool mark 75% of its comments as useful", "speaker": "Sonal Mahajan", "role": "Staff Engineer, Programming Systems team", "organization": "Uber", "aiVendor": false, "work": { "title": "uReview: Scalable, Trustworthy GenAI for Code Review at Uber", "type": "post", "date": "2025-08-12", "publisher": "Uber engineering blog", "url": "https://www.uber.com/us/en/blog/ureview/" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/uber-ureview-today-analyzes-over-90-weekly-65-2025-2" }, { "id": "openai-agentic-code-reviews-look-code-changes-series-2026", "text": "We're now seeing agentic code reviews that look at code changes through a series of different lenses. In the past, it would have been impractical for a cloud infrastructure engineer and a security engineer to review every single code change. With agents, that becomes possible.", "speaker": "Venkat Venkataramani", "role": "VP of Engineering, Applied Infra", "organization": "OpenAI", "aiVendor": true, "work": { "title": "Inside OpenAI's agentic software factory", "type": "press", "date": "2026-09-15", "publisher": "The Pragmatic Engineer", "locator": "2. Death of the IDE & pull requests", "url": "https://newsletter.pragmaticengineer.com/p/openai-software-factory" }, "grade": "C", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering", "security" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/openai-agentic-code-reviews-look-code-changes-series-2026" }, { "id": "warp-reviewer-agent-surfaces-issues-inconsistencies-human-2026", "text": "The reviewer agent surfaces issues and inconsistencies; the human makes the final call on quality and correctness.", "speaker": null, "role": null, "organization": "Warp", "aiVendor": true, "work": { "title": "Set up your software factory", "type": "docs", "date": "2026-10-28", "publisher": "Warp docs", "locator": "5. Human review: merge the PR", "url": "https://docs.warp.dev/guides/agent-workflows/set-up-a-software-factory" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/warp-reviewer-agent-surfaces-issues-inconsistencies-human-2026" }, { "id": "ramp-finds-right-human-reviewer-loop-2026", "text": "It finds the right human reviewer to loop in.", "speaker": "Geoff Charles", "role": "Chief Product Officer (CPO)", "organization": "Ramp", "aiVendor": false, "work": { "title": "What product looks like when coding is solved | Geoff Charles (Ramp CPO)", "type": "talk", "date": "2026-09-25", "publisher": "Lenny's Podcast (YouTube channel); Geoff Charles's talk at the Lenny and Friends Summit, recorded 2026-09-10", "url": "https://www.youtube.com/watch?v=ZG8Mf3P9xzI" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/ramp-finds-right-human-reviewer-loop-2026" }, { "id": "uber-growing-share-sessions-arent-initiated-humans-automated-2026", "text": "a growing share of sessions aren’t initiated by humans, but by automated managed agents handling code review, self-healing CI failures, completing E2E PRs with visual validation, triaging on-call alerts, debugging incoming bugs, and handling a variety of code maintenance tasks with human reviews/escalations.", "speaker": "Uday Kiran Medisetty", "role": "Distinguished Engineer", "organization": "Uber", "aiVendor": false, "work": { "title": "Running a Software Factory Efficiently at Uber Scale", "type": "post", "date": "2026-08-27", "publisher": "Uber engineering blog", "url": "https://www.uber.com/us/en/blog/efficient-software-factory/" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "bugs-and-incidents", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/uber-growing-share-sessions-arent-initiated-humans-automated-2026" }, { "id": "uber-if-autonomous-diff-coming-minion-want-give-2026", "text": "if this is an autonomous diff coming from Minion, we want to give a human reviewer some confidence that this diff has gone through a lot of self-improvement already", "speaker": "Uday Kiran Medisetty", "role": "Distinguished Engineer", "organization": "Uber", "aiVendor": false, "work": { "title": "Agentic SDLC at Uber: Building Blocks for Uber's Software Factory (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-21", "publisher": "AI Engineer (conference organizer) YouTube channel", "url": "https://www.youtube.com/watch?v=17-YSUHo6Lk" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/uber-if-autonomous-diff-coming-minion-want-give-2026" }, { "id": "shopify-mode-automation-pr-review-durable-session-woken-2026", "text": "Mode · Automation: PR review. Durable session, woken by an external system, often no human in the loop.", "speaker": "Burke Libbey", "role": "Principal Engineer", "organization": "Shopify", "aiVendor": false, "work": { "title": "Under the River", "type": "post", "date": "2026-05-28", "publisher": "Shopify engineering blog", "url": "https://shopify.engineering/under-the-river" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "coding-agents", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/shopify-mode-automation-pr-review-durable-session-woken-2026" }, { "id": "cloudflare-over-last-30-days-100-ai-code-2026", "text": "Over the last 30 days: 100% AI code reviewer coverage across all repos on our standard CI pipeline.", "speaker": null, "role": null, "organization": "Cloudflare", "aiVendor": true, "work": { "title": "The AI engineering stack we built internally, on the platform we ship", "type": "post", "date": "2026-04-20", "publisher": "Cloudflare blog", "url": "https://blog.cloudflare.com/internal-ai-engineering-stack/" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/cloudflare-over-last-30-days-100-ai-code-2026" }, { "id": "hubspot-code-reviewed-over-50-000-pull-requests-2026", "text": "code reviewed over 50,000 pull requests authored by humans", "speaker": null, "role": null, "organization": "HubSpot", "aiVendor": true, "work": { "title": "Cloud Coding Agents at HubSpot", "type": "post", "date": "2026-01-23", "publisher": "HubSpot product blog", "url": "https://product.hubspot.com/blog/cloud-coding-agents-at-hubspot" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/hubspot-code-reviewed-over-50-000-pull-requests-2026" }, { "id": "anthropic-labs-bottlenecked-on-review-human-understanding-2026", "text": "We are definitely still bottlenecked on reviews, especially for things that are touching some architecture pieces. And it's actually more subtle than just being bottlenecked on review, … we can carve out time differently. It's … bottlenecked on human ability to even fully conceptualize what we're doing.", "speaker": "Mike Krieger", "role": "Head of Anthropic Labs", "organization": "Anthropic", "aiVendor": true, "work": { "title": "How Anthropic Builds: Lessons from Labs", "type": "interview", "date": "2026-07-02", "publisher": "AI Engineer (YouTube)", "locator": "10:13", "url": "https://www.youtube.com/watch?v=qqrk7CtkuIw" }, "grade": "C", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": "Answering a question from the host on whether code review is a bottleneck and whether code could be merged without a person looking at it.", "url": "https://the-ai-record.com/passages/anthropic-labs-bottlenecked-on-review-human-understanding-2026" }, { "id": "anthropic-labs-talk-claude-about-code-instead-reading-every-line-2026", "text": "actually talked to Claude about the code and say, all right, these are the questions that I would have. Can you go investigate it? So it is kind of … code review, but still human-driven, and for the really important ones. And for the ones that are cosmetic visual changes, it's much more like we'll fix forward if we need to fix forward.", "speaker": "Mike Krieger", "role": "Head of Anthropic Labs", "organization": "Anthropic", "aiVendor": true, "work": { "title": "How Anthropic Builds: Lessons from Labs", "type": "interview", "date": "2026-07-02", "publisher": "AI Engineer (YouTube)", "locator": "11:13", "url": "https://www.youtube.com/watch?v=qqrk7CtkuIw" }, "grade": "C", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": "Continuing his answer on review. The sentence before this passage, on not reviewing every line of a pull request, is not quoted because the two passes disagreed on its wording.", "url": "https://the-ai-record.com/passages/anthropic-labs-talk-claude-about-code-instead-reading-every-line-2026" }, { "id": "anthropic-pull-requests-sent-with-artifact-explaining-intent-tradeoffs-2026", "text": "So one of the reasons we built … artifacts that we shipped a couple of weeks ago was partially for that, which is you would send somebody a PR, and then they'd be like, I don't know, man, this is like 2000 lines of code. It looks like code to me. And what we started doing instead is sharing much more like, here's a … artifact. Here's the explanation. Here's the intention of the change. Here's the trade-offs that were made.", "speaker": "Mike Krieger", "role": "Head of Anthropic Labs", "organization": "Anthropic", "aiVendor": true, "work": { "title": "How Anthropic Builds: Lessons from Labs", "type": "interview", "date": "2026-07-02", "publisher": "AI Engineer (YouTube)", "locator": "10:30", "url": "https://www.youtube.com/watch?v=qqrk7CtkuIw" }, "grade": "C", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "developer-productivity" ], "functions": [ "engineering" ], "figures": [], "context": "Continuing his answer on review bottlenecks. The product name before 'artifacts' is rendered unclearly in the recording's transcripts, so the passage is cut around it.", "url": "https://the-ai-record.com/passages/anthropic-pull-requests-sent-with-artifact-explaining-intent-tradeoffs-2026" }, { "id": "greptile-almost-fifth-prs-merged-without-human-review-2026", "text": "You get a pretty high degree of confidence on merge. Today, almost a fifth of all the pull … are merged without any human review or without any human testing. … that is a number that we care a lot about and want to bring higher and higher, of course, within the guardrails of producing really high quality code.", "speaker": "Daksh Gupta", "role": "Co-founder", "organization": "Greptile", "aiVendor": true, "work": { "title": "AI-Generated Code Is Already Competing With Human Code", "type": "talk", "date": "2026-09-27", "publisher": "AI Engineer (YouTube)", "locator": "11:50", "url": "https://www.youtube.com/watch?v=474j-n1Ltxc" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "autonomy-and-oversight", "code-review" ], "functions": [ "engineering" ], "figures": [ { "metric": "share-prs-merged-without-human-review", "value": 0.2, "unit": "share", "period": "as of the talk (almost a fifth)", "definition": "pull requests that Greptile reviews that are merged without any human review or human testing, per the speaker (unit of work: PRs)", "level": "human-out" } ], "context": null, "url": "https://the-ai-record.com/passages/greptile-almost-fifth-prs-merged-without-human-review-2026" }, { "id": "greptile-little-difference-revert-rate-people-versus-agents-2026", "text": "So there didn't seem to be very big difference between the rate at which pull requests were reverted from people versus agents in my study.", "speaker": "Daksh Gupta", "role": "Co-founder", "organization": "Greptile", "aiVendor": true, "work": { "title": "AI-Generated Code Is Already Competing With Human Code", "type": "talk", "date": "2026-09-27", "publisher": "AI Engineer (YouTube)", "locator": "06:05", "url": "https://www.youtube.com/watch?v=474j-n1Ltxc" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/greptile-little-difference-revert-rate-people-versus-agents-2026" }, { "id": "greptile-swarm-agents-every-pull-request-sandbox-2026", "text": "every time there's a pull request, we have a swarm of agents. And they go and look at every file that's changed, every file that's related to the files that change to figure out if there's bugs. And it also spins up your code in a sandbox, … dependencies, spins up local host, clicks around to try to break things, mocks inputs, and everything else to figure out if the code is broken.", "speaker": "Daksh Gupta", "role": "Co-founder", "organization": "Greptile", "aiVendor": true, "work": { "title": "AI-Generated Code Is Already Competing With Human Code", "type": "talk", "date": "2026-09-27", "publisher": "AI Engineer (YouTube)", "locator": "00:27", "url": "https://www.youtube.com/watch?v=474j-n1Ltxc" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "testing" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/greptile-swarm-agents-every-pull-request-sandbox-2026" }, { "id": "greptile-three-out-four-agents-fewer-p0s-than-humans-2026", "text": "So then I started tracking the counts of P0s, P1s, and P2s in all of these changes. And interestingly, once again, there was not that much of a difference. In fact, three out of the four agents that we tested performed better than humans in terms of the rate at which … were producing P0s. They were producing fewer P0s than humans were. Similarly, the case with P1s and P2s. Broadly speaking, human-generated PRs were about equal in quality to … PRs based on this data.", "speaker": "Daksh Gupta", "role": "Co-founder", "organization": "Greptile", "aiVendor": true, "work": { "title": "AI-Generated Code Is Already Competing With Human Code", "type": "talk", "date": "2026-09-27", "publisher": "AI Engineer (YouTube)", "locator": "07:07", "url": "https://www.youtube.com/watch?v=474j-n1Ltxc" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/greptile-three-out-four-agents-fewer-p0s-than-humans-2026" }, { "id": "greptile-very-little-difference-review-rounds-human-ai-prs-2026", "text": "And so I started tracking the number of iterations, the number of review rounds between when the pull request was opened and when it was merged. Sure enough, very little difference. … Once again, very little if any statistical difference between human-generated and AI-generated pull requests in terms of how many iterations before they're ready to merge.", "speaker": "Daksh Gupta", "role": "Co-founder", "organization": "Greptile", "aiVendor": true, "work": { "title": "AI-Generated Code Is Already Competing With Human Code", "type": "talk", "date": "2026-09-27", "publisher": "AI Engineer (YouTube)", "locator": "07:57", "url": "https://www.youtube.com/watch?v=474j-n1Ltxc" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "measuring-impact" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/greptile-very-little-difference-review-rounds-human-ai-prs-2026" }, { "id": "figma-pr-description-starts-with-hand-written-ai-description-after-2026", "text": "we've decided we always, every PR description will start with something like that, something that I wrote by hand … could be very short to describe what this is doing and what this is doing.", "speaker": "Eyal Blum", "role": "Software Engineer", "organization": "Figma", "aiVendor": true, "work": { "title": "How to Get Your Org to Adopt Coding Agents (Without Shipping Garbage)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "13:41", "url": "https://www.youtube.com/watch?v=5Bn0xro2ol8" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "ai-across-the-company" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/figma-pr-description-starts-with-hand-written-ai-description-after-2026" }, { "id": "boundary-dependency-graph-cli-invariants-architecture-unchanged-months-2026", "text": "This tool basically visualizes our dependency graph internally with some external dependencies as well, … But what's more interesting is we can go build CLI tools that guarantee that certain invariants can't be broken. And what this does is when Claude builds a new package or adds a dependency that's leaky, we now have CI/CD changing or a simple git commit history that tells us exactly where things break. And by this, we're actually able to make our architecture change. We haven't changed our architecture in the last three or four months.", "speaker": "Vaibhav Gupta", "role": "Co-founder and Chief Executive", "organization": "Boundary (BoundaryML)", "aiVendor": true, "work": { "title": "Fighting Slop with Slop", "type": "talk", "date": "2026-07-31", "publisher": "AI Engineer (YouTube)", "locator": "4:05", "url": "https://www.youtube.com/watch?v=AMiyLItEtLA" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": "Gupta is describing how the team keeps the BAML code base from drifting when engineers do not review each other's code.", "url": "https://the-ai-record.com/passages/boundary-dependency-graph-cli-invariants-architecture-unchanged-months-2026" }, { "id": "boundary-no-code-reviews-parallel-no-ai-standardization-2026", "text": "We do no code reviews. We require every engineer to work on things in parallel, and we have no standardization how people do AI.", "speaker": "Vaibhav Gupta", "role": "Co-founder and Chief Executive", "organization": "Boundary (BoundaryML)", "aiVendor": true, "work": { "title": "Fighting Slop with Slop", "type": "talk", "date": "2026-07-31", "publisher": "AI Engineer (YouTube)", "locator": "0:26", "url": "https://www.youtube.com/watch?v=AMiyLItEtLA" }, "grade": "B", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight", "developer-productivity" ], "functions": [ "engineering" ], "figures": [], "context": "Opening of a conference talk by the chief executive of Boundary, the company that builds the BAML programming language, on how its team works.", "url": "https://the-ai-record.com/passages/boundary-no-code-reviews-parallel-no-ai-standardization-2026" }, { "id": "cognition-vibe-coding-no-review-experiment-two-weeks-2026", "text": "I think we found that the state-of-the-art in December was you can probably, run this for about two weeks. By the end of those two weeks, you’d find that, hey, you want to, change the color of a button. Well, it turns out this button is implemented in, 10 different places, and they, have All these different variations, and oh, you forgot one of them, and actually it’s a slightly different color in one spot.", "speaker": "Walden Yan", "role": "Co-founder and Chief Product Officer", "organization": "Cognition", "aiVendor": true, "work": { "title": "The Age of Async Agents: Cognition's Walden Yan and OpenInspect's Cole Murray", "type": "podcast", "date": "2026-05-28", "publisher": "Latent Space", "locator": "00:44:21", "url": "https://latent.space/p/cognition" }, "grade": "C", "verification": { "status": "verified", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": "Yan was describing an internal experiment building real products by auto-merging AI-written code with no code review, to see how many weeks it could run before the codebase became unmanageable.", "url": "https://the-ai-record.com/passages/cognition-vibe-coding-no-review-experiment-two-weeks-2026" }, { "id": "uber-back-2024-we-were-seeing-engineers-2026", "text": "Back in 2024, we were seeing that engineers would get their first review within three hours. Now, in 2026, that has grown to nine hours, in addition to all of the volume changes. So in short, code review is now the bottleneck that we are running into.", "speaker": "Will Bond", "role": "Staff Engineer, AI Foundations team, Development Platform organization", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "01:05", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "developer-productivity" ], "functions": [ "engineering" ], "figures": [ { "metric": "first-review-time-2024", "value": 3, "unit": "hours", "period": "2024", "definition": "time for an engineer to get a first review on a pull request, as Uber tracks it", "level": "n/a" }, { "metric": "first-review-time-2026", "value": 9, "unit": "hours", "period": "2026", "definition": "time for an engineer to get a first review on a pull request, as Uber tracks it", "level": "n/a" } ], "context": "Will Bond is describing a metric Uber tracks, the time to a first review of a pull request.", "url": "https://the-ai-record.com/passages/uber-back-2024-we-were-seeing-engineers-2026" }, { "id": "uber-currently-we-still-have-humans-approving-2026", "text": "Currently, we still have humans approving the code, but we see a short path in the near future to a percentage of our code landing automatically, having automatic approvals, right?", "speaker": "Will Bond", "role": "Staff Engineer, AI Foundations team, Development Platform organization", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "11:39", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "autonomy-and-oversight", "code-review" ], "functions": [ "engineering" ], "figures": [], "context": "Will Bond is describing where Uber expects code review to go as it moves to an agentic software development life cycle.", "url": "https://the-ai-record.com/passages/uber-currently-we-still-have-humans-approving-2026" }, { "id": "uber-one-biggest-learnings-process-like-model-2026", "text": "One of the biggest learnings in this process was like the model doesn't know that it's wrong. It always confidently says 100% sure that, yeah, this is the review for your code. Go ahead. But we saw that, no, it actually needs a lot of guidance from the teams, because each team has its own style guide, its own patterns or … anti-patterns that they want to look for.", "speaker": "Ameya Ketkar", "role": "Speaker, AI Engineer World's Fair 2026 (job title not stated)", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "06:35", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "coding-agents" ], "functions": [ "engineering" ], "figures": [], "context": "Ameya Ketkar is describing what Uber's team learned while adding observability and evaluation to its uReview code-review agent.", "url": "https://the-ai-record.com/passages/uber-one-biggest-learnings-process-like-model-2026" }, { "id": "uber-one-thing-we-learned-actually-writing-2026", "text": "One thing that we learned is that actually writing the skill was very easy, like teams just very quickly wrote a skill by asking Claude to write one, go over my previous PR reviews and write me a skill, but the hard part was how to run these skills at scale with consistent quality and low cost, and that required a lot of iterations, not only from the [uReview] team side, but also … for each team who was trying to write these rules.", "speaker": "Ameya Ketkar", "role": "Speaker, AI Engineer World's Fair 2026 (job title not stated)", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "09:44", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "internal-tools" ], "functions": [ "engineering" ], "figures": [], "context": "Ameya Ketkar is describing how Uber's teams wrote their own code-review skills for uReview.", "url": "https://the-ai-record.com/passages/uber-one-thing-we-learned-actually-writing-2026" }, { "id": "uber-rather-than-killing-outer-loop-i-2026", "text": "But rather than killing the outer loop, I think that we believe and the industry has just started to really kind of coalesce on this idea that … really expanding the outer loop. Rather than removing humans from the code review process, we are moving their responsibilities up a layer.", "speaker": "Will Bond", "role": "Staff Engineer, AI Foundations team, Development Platform organization", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "13:43", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "autonomy-and-oversight", "code-review" ], "functions": [ "engineering" ], "figures": [], "context": "Will Bond is closing the talk on what happens to human reviewers as agents write and review more of the code.", "url": "https://the-ai-record.com/passages/uber-rather-than-killing-outer-loop-i-2026" }, { "id": "uber-uber-spent-time-invest-building-house-2026", "text": "Uber spent the time to invest in building an in-house solution due to some of the constraints that we have. One of those is we currently use [Phabricator] and have for a long time and are in the process of migrating to GitHub. Most of the solutions do not provide support for [Phabricator].", "speaker": "Will Bond", "role": "Staff Engineer, AI Foundations team, Development Platform organization", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "01:22", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "internal-tools" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/uber-uber-spent-time-invest-building-house-2026" }, { "id": "uber-ureview-does-around-25-000-comments-2026", "text": "[uReview] does … around 25,000 comments a week, and we get 10% of them actually get some feedback … We also saw that the overall address rate was around 67%, and almost three quarters of the high severity issues were usually addressed by the developers", "speaker": "Ameya Ketkar", "role": "Speaker, AI Engineer World's Fair 2026 (job title not stated)", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "10:13", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "measuring-impact" ], "functions": [ "engineering" ], "figures": [ { "metric": "ai-review-comments", "value": 25000, "unit": "comments", "period": "per week", "definition": "comments posted by uReview, Uber's AI code-review system", "level": "assisted" }, { "metric": "review-comment-address-rate", "value": 67, "unit": "percent", "period": "as of the talk", "definition": "share of uReview comments that the developer went and addressed, overall", "level": "assisted" } ], "context": "Ameya Ketkar is reporting results for uReview, Uber's code-review system. The address rate is how often a developer acts on a comment, as he defined it earlier in the talk.", "url": "https://the-ai-record.com/passages/uber-ureview-does-around-25-000-comments-2026" }, { "id": "uber-we-saw-against-like-very-naive-2026", "text": "we saw that against like a very naive implementation, our costs were down by 60%, and our quality and our accuracy was up by around 70%.", "speaker": "Ameya Ketkar", "role": "Speaker, AI Engineer World's Fair 2026 (job title not stated)", "organization": "Uber", "aiVendor": false, "work": { "title": "Building uReview, Uber's Multi-Agent Code Review Engine (talk, AI Engineer World's Fair 2026)", "type": "talk", "date": "2026-08-28", "publisher": "AI Engineer (YouTube)", "locator": "10:50", "url": "https://www.youtube.com/watch?v=EL123UNokkI" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "ai-cost-and-spend", "measuring-impact" ], "functions": [ "engineering" ], "figures": [ { "metric": "ai-review-cost-change", "value": -60, "unit": "percent", "period": "against a naive first implementation", "definition": "cost of running uReview, compared with a very naive implementation, as Ketkar describes it", "level": "n/a" }, { "metric": "ai-review-quality-change", "value": 70, "unit": "percent", "period": "against a naive first implementation", "definition": "quality and accuracy of uReview, compared with a very naive implementation, as Ketkar describes it; the talk does not say how this is measured", "level": "n/a" } ], "context": "Ameya Ketkar is summarizing the effect of the observability and evaluation work on uReview, Uber's code-review system.", "url": "https://the-ai-record.com/passages/uber-we-saw-against-like-very-naive-2026" }, { "id": "cloudflare-engineers-dont-trust-ai-opinions-clear-findings-2026", "text": "engineers don't trust AI opinions. They just trust clear findings tied to standards that we … agreed upon. … it stopped sounding more like, maybe this is a problem and started sounding more like this violates a standard that we actually care about.", "speaker": "Rajesh Bhatia", "role": "Senior Director of Engineering, Cloudflare", "organization": "Cloudflare", "aiVendor": true, "work": { "title": "From Assisted to Delegated: Cloudflare's AI Engineering Stack (interview, Background Agents Summit 2026)", "type": "talk", "date": "2026-05-06", "publisher": "Ona (YouTube)", "locator": "26:40", "url": "https://www.youtube.com/watch?v=MbLdrAZFQRs" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": "Describing early misses with Cloudflare's internal AI code reviewer, before the team tied its findings to machine-readable engineering standards.", "url": "https://the-ai-record.com/passages/cloudflare-engineers-dont-trust-ai-opinions-clear-findings-2026" }, { "id": "cloudflare-network-firewall-team-audited-rampartd-multi-agent-2026", "text": "Our Network Firewall team audited rampartd using a multi-agent consensus process where every requirement was scored COMPLIANT, PARTIAL, or NON-COMPLIANT with specific violation details and remediation steps reducing what previously required weeks of manual work to a structured, repeatable process.", "speaker": null, "role": null, "organization": "Cloudflare", "aiVendor": true, "work": { "title": "The AI engineering stack we built internally, on the platform we ship", "type": "post", "date": "2026-04-20", "publisher": "Cloudflare blog", "locator": "Act 3, \"Engineering Codex: engineering standards as agent skills\"", "url": "https://blog.cloudflare.com/internal-ai-engineering-stack/" }, "grade": "B", "verification": { "status": "reported" }, "topics": [ "code-review", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/cloudflare-network-firewall-team-audited-rampartd-multi-agent-2026" }, { "id": "klarna-june-2025-80-engineers-data-scientists-connected-2025", "text": "As of June 2025, 80% of our engineers and data scientists connected to their work software an AI copilot that can create and review code", "speaker": null, "role": null, "organization": "Klarna", "aiVendor": false, "work": { "title": "SEC filing (the company's own regulatory report)", "type": "filing", "date": "2025-09-08", "publisher": "SEC filing (the company's own regulatory report)", "url": "https://www.sec.gov/Archives/edgar/data/2003292/000200329225000036/klarnagroupplcf-1a4.htm" }, "grade": "A", "verification": { "status": "verified", "reviewedOn": "2026-10-31" }, "topics": [ "code-review" ], "functions": [ "engineering" ], "figures": [], "context": null, "url": "https://the-ai-record.com/passages/klarna-june-2025-80-engineers-data-scientists-connected-2025" } ], "notice": [ "Every passage is reproduced word for word from the cited source.", "Check the verification status: \"reported\" means not yet confirmed by an independent second reader.", "Organizations flagged aiVendor sell AI tooling and have a reason to publish flattering numbers.", "An empty result means the archive holds nothing on it, not that nobody has said anything." ] }