{ "apiVersion": "v1", "topic": { "id": "reading-agent-code", "title": "Do engineers still read the code agents write?", "group": "Reviewing agent code", "summary": "Whether people read agent-written code line by line, skim it, ask the agent to explain it, or skip review, and who answers for it.", "passages": 18, "url": "https://the-ai-record.com/topics/reading-agent-code", "json": "https://the-ai-record.com/api/v1/topics/reading-agent-code.json" }, "passages": [ { "id": "posthog-agents-writing-about-70-our-prs-2026", "text": "PostHog is just a little further than that with agents writing about 70% of our PRs, but humans still skim at least 80% of those.", "context": null, "speaker": "Jina Yoon", "speakerId": "jina-yoon", "speakerUrl": "https://the-ai-record.com/people/jina-yoon", "speakerLevel": "practitioner", "role": "Author, PostHog newsletter", "voice": "person", "organization": "PostHog", "organizationId": "posthog", "organizationUrl": "https://the-ai-record.com/organizations/posthog", "aiVendor": true, "work": { "title": "Can software factories actually work?", "type": "post", "date": "2026-08-11", "publisher": "PostHog newsletter", "locator": "What is a software factory?", "url": "https://posthog.com/newsletter/software-factories" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "coding-agents", "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [ { "metric": "agent-written-pr-share", "value": 0.7, "unit": "share", "definition": "Share of PostHog's pull requests the passage says agents write ('about 70%')", "level": "human-approves" } ], "url": "https://the-ai-record.com/passages/posthog-agents-writing-about-70-our-prs-2026", "json": "https://the-ai-record.com/api/v1/passages/posthog-agents-writing-about-70-our-prs-2026.json" }, { "id": "strongdm-zero-hand-written-code-zero-traditional-review-2026", "text": "Given zero hand-written code and zero traditional review, we required a system that could: Grow from cascades of natural-language specifications Be validated automatically without semantic inspection of source", "context": null, "speaker": null, "speakerId": null, "speakerUrl": null, "speakerLevel": null, "role": null, "voice": "organization", "organization": "StrongDM", "organizationId": "strongdm", "organizationUrl": "https://the-ai-record.com/organizations/strongdm", "aiVendor": false, "work": { "title": "Techniques: The Validation Constraint (StrongDM Software Factory)", "type": "docs", "date": "2026-02", "publisher": "StrongDM AI", "locator": "The Validation Constraint", "url": "https://factory.strongdm.ai/techniques" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-07" }, "topics": [ "ai-written-code", "code-review", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/strongdm-zero-hand-written-code-zero-traditional-review-2026", "json": "https://the-ai-record.com/api/v1/passages/strongdm-zero-hand-written-code-zero-traditional-review-2026.json" }, { "id": "airbnb-every-engineer-must-able-explain-even-if-2026", "text": "every engineer must be able to explain, even if an AI generated the PR, must be able to explain what they built.", "context": null, "speaker": "Ahmad Al-Dahle", "speakerId": "ahmad-al-dahle", "speakerUrl": "https://the-ai-record.com/people/ahmad-al-dahle", "speakerLevel": "executive", "role": "Chief Technology Officer", "voice": "person", "organization": "Airbnb", "organizationId": "airbnb", "organizationUrl": "https://the-ai-record.com/organizations/airbnb", "aiVendor": false, "work": { "title": "Latent Space (independent AI engineering publication)", "type": "interview", "date": "2026-10-02", "publisher": "Latent Space (independent AI engineering publication)", "url": "https://www.latent.space/p/airbnb" }, "grade": "C", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "coding-agents", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/airbnb-every-engineer-must-able-explain-even-if-2026", "json": "https://the-ai-record.com/api/v1/passages/airbnb-every-engineer-must-able-explain-even-if-2026.json" }, { "id": "strongdm-we-built-software-factory-non-interactive-development-2026", "text": "We built a Software Factory: non-interactive development where specs + scenarios drive agents that write code, run harnesses, and converge without human review.", "context": null, "speaker": null, "speakerId": null, "speakerUrl": null, "speakerLevel": null, "role": null, "voice": "organization", "organization": "StrongDM", "organizationId": "strongdm", "organizationUrl": "https://the-ai-record.com/organizations/strongdm", "aiVendor": false, "work": { "title": "How StrongDM's AI team build serious software without even looking at the code", "type": "press", "date": "2026-02-07", "publisher": "Simon Willison's Weblog", "url": "https://simonwillison.net/2026/feb/7/software-factory/" }, "grade": "C", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-06" }, "topics": [ "coding-agents", "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/strongdm-we-built-software-factory-non-interactive-development-2026", "json": "https://the-ai-record.com/api/v1/passages/strongdm-we-built-software-factory-non-interactive-development-2026.json" }, { "id": "anthropic-claude-code-merged-prs-doubled-reverts-fell-2026", "text": "From June to September, merged pull requests to Claude Code's codebase more than doubled while the share that had to be reverted fell by more than 40 percent. We didn't remove review. But we did move most of the reading to machines, optimized for human time and attention, and invested heavily in making it fast and easy to undo mistakes.", "context": null, "speaker": null, "speakerId": null, "speakerUrl": null, "speakerLevel": null, "role": null, "voice": "organization", "organization": "Anthropic", "organizationId": "anthropic", "organizationUrl": "https://the-ai-record.com/organizations/anthropic", "aiVendor": true, "work": { "title": "How the Claude Code team does code review", "type": "post", "date": "2026-10-15", "publisher": "claude.dev Blog (Anthropic)", "locator": "Introduction", "url": "https://claude.dev/blog/solving-code-review-at-anthropic/" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-07" }, "topics": [ "reading-agent-code", "code-review" ], "functions": [ "engineering" ], "figures": [ { "metric": "output-change", "value": 2, "unit": "ratio", "period": "June to September 2026", "definition": "Merged pull requests to Claude Code's codebase, September against June (\"more than doubled\"; lower bound)", "level": "unclear" }, { "metric": "revert-rate-reduction", "value": 0.4, "unit": "share", "period": "June to September 2026", "definition": "Fall in the share of merged pull requests to Claude Code's codebase that had to be reverted (\"more than 40 percent\"; lower bound)", "level": "unclear" } ], "url": "https://the-ai-record.com/passages/anthropic-claude-code-merged-prs-doubled-reverts-fell-2026", "json": "https://the-ai-record.com/api/v1/passages/anthropic-claude-code-merged-prs-doubled-reverts-fell-2026.json" }, { "id": "anthropic-human-read-code-imprecise-proxy-2026", "text": "We think \"did a human read the code?\" is an imprecise proxy for \"what could go wrong with this change, and which check—automated or human—is the right one to catch it?\"", "context": null, "speaker": null, "speakerId": null, "speakerUrl": null, "speakerLevel": null, "role": null, "voice": "organization", "organization": "Anthropic", "organizationId": "anthropic", "organizationUrl": "https://the-ai-record.com/organizations/anthropic", "aiVendor": true, "work": { "title": "How the Claude Code team does code review", "type": "post", "date": "2026-10-15", "publisher": "claude.dev Blog (Anthropic)", "locator": "Rethinking the purpose of code review", "url": "https://claude.dev/blog/solving-code-review-at-anthropic/" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-07" }, "topics": [ "reading-agent-code", "code-review" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/anthropic-human-read-code-imprecise-proxy-2026", "json": "https://the-ai-record.com/api/v1/passages/anthropic-human-read-code-imprecise-proxy-2026.json" }, { "id": "anthropic-people-still-own-risky-code-and-outcome-2026", "text": "Pulling a senior engineer out of deep work to catch a routine mistake costs far more than an automated review. People on the team are still responsible for: … Reviewing risky code such as billing, the auto-updater and CI workflows. … Owning the outcome. A person is accountable for every change that ships, whoever typed it.", "context": "The two cuts drop three items from the post's list of what people stay responsible for: deciding whether to build a change, reviewing architecture that sets a pattern others will copy, and reading the evidence the checks produce.", "speaker": null, "speakerId": null, "speakerUrl": null, "speakerLevel": null, "role": null, "voice": "organization", "organization": "Anthropic", "organizationId": "anthropic", "organizationUrl": "https://the-ai-record.com/organizations/anthropic", "aiVendor": true, "work": { "title": "How the Claude Code team does code review", "type": "post", "date": "2026-10-15", "publisher": "claude.dev Blog (Anthropic)", "locator": "Where humans stay in the loop", "url": "https://claude.dev/blog/solving-code-review-at-anthropic/" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-07" }, "topics": [ "reading-agent-code", "autonomy-and-oversight" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/anthropic-people-still-own-risky-code-and-outcome-2026", "json": "https://the-ai-record.com/api/v1/passages/anthropic-people-still-own-risky-code-and-outcome-2026.json" }, { "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.", "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.", "speaker": "Vaibhav Gupta", "speakerId": "vaibhav-gupta", "speakerUrl": "https://the-ai-record.com/people/vaibhav-gupta", "speakerLevel": "executive", "role": "Co-founder and Chief Executive", "voice": "person", "organization": "Boundary (BoundaryML)", "organizationId": "boundary", "organizationUrl": "https://the-ai-record.com/organizations/boundary", "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", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight", "developer-productivity", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/boundary-no-code-reviews-parallel-no-ai-standardization-2026", "json": "https://the-ai-record.com/api/v1/passages/boundary-no-code-reviews-parallel-no-ai-standardization-2026.json" }, { "id": "humanlayer-loops-still-read-the-code-2026", "text": "So today I want to talk about what I think works in the real world and what we've started doing at HumanLayer, which, to be clear, is still building loops, right? I think loops are super powerful, but we can design loops and still read the code. In fact, we can design loops that make it easier to read the code because the loops are making the code better.", "context": "Mistele is opening the part of his talk about how his company builds agent loops, after arguing that loops which produce unreadable pull requests are the wrong design.", "speaker": "Kyle Mistele", "speakerId": "kyle-mistele", "speakerUrl": "https://the-ai-record.com/people/kyle-mistele", "speakerLevel": "executive", "role": "Co-founder, HumanLayer", "voice": "person", "organization": "HumanLayer", "organizationId": "humanlayer", "organizationUrl": "https://the-ai-record.com/organizations/humanlayer", "aiVendor": true, "work": { "title": "Loop Engineering from first principles", "type": "talk", "date": "2026-07-25", "publisher": "AI Engineer (YouTube)", "locator": "4:03", "url": "https://www.youtube.com/watch?v=xIt_mTQp6mY" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-03" }, "topics": [ "coding-agents", "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/humanlayer-loops-still-read-the-code-2026", "json": "https://the-ai-record.com/api/v1/passages/humanlayer-loops-still-read-the-code-2026.json" }, { "id": "humanlayer-small-incremental-loops-two-to-three-times-faster-owning-architecture-2026", "text": "And so my advice is … instead of trying to automate everything end-to-end, build these small incremental loops throughout your system, and you will wake up one day and you will be moving two to three times faster while still being able to read the code, while still owning the architecture.", "context": "Horthy is answering the moderator's question on how to reach fully autonomous software factories if two to three times is what coding agents give most engineers.", "speaker": "Dex Horthy", "speakerId": "dex-horthy", "speakerUrl": "https://the-ai-record.com/people/dex-horthy", "speakerLevel": "executive", "role": "CEO, HumanLayer", "voice": "person", "organization": "HumanLayer", "organizationId": "humanlayer", "organizationUrl": "https://the-ai-record.com/organizations/humanlayer", "aiVendor": true, "work": { "title": "The Great Loops Debate", "type": "talk", "date": "2026-07-17", "publisher": "AI Engineer (YouTube)", "locator": "48:33", "url": "https://www.youtube.com/watch?v=c35YoMdnI78" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-04" }, "topics": [ "developer-productivity", "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [ { "metric": "speedup-low", "value": 2, "unit": "x", "period": "as the speaker states it", "definition": "low end of the speed-up he says small incremental loops give while the engineer still reads the code", "level": "human-approves" }, { "metric": "speedup-high", "value": 3, "unit": "x", "period": "as the speaker states it", "definition": "high end of the speed-up he says small incremental loops give while the engineer still reads the code", "level": "human-approves" } ], "url": "https://the-ai-record.com/passages/humanlayer-small-incremental-loops-two-to-three-times-faster-owning-architecture-2026", "json": "https://the-ai-record.com/api/v1/passages/humanlayer-small-incremental-loops-two-to-three-times-faster-owning-architecture-2026.json" }, { "id": "stripe-while-theyre-human-reviewed-they-contain-no-2026", "text": "while they're human-reviewed, they contain no human-written code.", "context": null, "speaker": "Alistair Gray", "speakerId": "alistair-gray", "speakerUrl": "https://the-ai-record.com/people/alistair-gray", "speakerLevel": "practitioner", "role": "Software engineer, Leverage team", "voice": "person", "organization": "Stripe", "organizationId": "stripe", "organizationUrl": "https://the-ai-record.com/organizations/stripe", "aiVendor": false, "work": { "title": "Minions: Stripe’s one-shot, end-to-end coding agents", "type": "post", "date": "2026-02-09", "publisher": "Stripe developer blog", "url": "https://stripe.dev/blog/minions-stripes-one-shot-end-to-end-coding-agents" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "coding-agents", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/stripe-while-theyre-human-reviewed-they-contain-no-2026", "json": "https://the-ai-record.com/api/v1/passages/stripe-while-theyre-human-reviewed-they-contain-no-2026.json" }, { "id": "vercel-read-the-code-asks-ai-to-explain-back-engineering-manager-2026", "text": "Now, you asked specifically, do I write the code? I read … and what this says, I'm doing more and more and more, is asking the AI to explain back what it did to me, which is a little bit of the job I've compared it to being an engineering manager.", "context": "Rauch answers the host's question 'Are you still coding?'", "speaker": "Guillermo Rauch", "speakerId": "guillermo-rauch", "speakerUrl": "https://the-ai-record.com/people/guillermo-rauch", "speakerLevel": "executive", "role": "Founder and CEO of Vercel", "voice": "person", "organization": "Vercel", "organizationId": "vercel", "organizationUrl": "https://the-ai-record.com/organizations/vercel", "aiVendor": true, "work": { "title": "60% of Vercel Deployments Are Now Shipped by AI Agents", "type": "interview", "date": "2026-10-16", "publisher": "Sourcery with Molly O'Shea (YouTube)", "locator": "12:17", "url": "https://www.youtube.com/watch?v=WuorvLfAFmA" }, "grade": "C", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-06" }, "topics": [ "autonomy-and-oversight", "coding-agents", "reading-agent-code", "engineer-role" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/vercel-read-the-code-asks-ai-to-explain-back-engineering-manager-2026", "json": "https://the-ai-record.com/api/v1/passages/vercel-read-the-code-asks-ai-to-explain-back-engineering-manager-2026.json" }, { "id": "uber-still-require-engineer-review-every-pr-green-lane-2026", "text": "right now we still require an engineer to review every PR … goes in. … Once the more confidence we get in our risk scoring and we can also back test it with years of data, right? We have all of this data and the more confidence we have in our AI code reviews, then maybe for some class of low risk PRs, we could enable, we call it green lane like auto landing.", "context": "Answering the host's question whether human code review will always exist.", "speaker": "Uday Kiran Medisetty", "speakerId": "uday-kiran-medisetty", "speakerUrl": "https://the-ai-record.com/people/uday-kiran-medisetty", "speakerLevel": "practitioner", "role": "Distinguished Engineer, Uber", "voice": "person", "organization": "Uber", "organizationId": "uber", "organizationUrl": "https://the-ai-record.com/organizations/uber", "aiVendor": false, "work": { "title": "Uber's AI infra, from the inside", "type": "podcast", "date": "2026-10-14", "publisher": "The Pragmatic Engineer (YouTube)", "locator": "59:37", "url": "https://www.youtube.com/watch?v=MDUYSMJmi8A" }, "grade": "C", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-07" }, "topics": [ "ai-approves-code", "code-review", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/uber-still-require-engineer-review-every-pr-green-lane-2026", "json": "https://the-ai-record.com/api/v1/passages/uber-still-require-engineer-review-every-pr-green-lane-2026.json" }, { "id": "shopify-you-just-tell-ai-go-nuts-makes-2026", "text": "You just tell the AI to go nuts. It makes a pull request. You just say, “Yep, that’s good.” You don’t really read it, and now it has to be reviewed by your colleagues.", "context": null, "speaker": "Tobi Lütke", "speakerId": "tobi-lutke", "speakerUrl": "https://the-ai-record.com/people/tobi-lutke", "speakerLevel": "executive", "role": "Co-founder and Chief Executive Officer", "voice": "person", "organization": "Shopify", "organizationId": "shopify", "organizationUrl": "https://the-ai-record.com/organizations/shopify", "aiVendor": false, "work": { "title": "The Knowledge Project (podcast by Shane Parrish of Farnam Street, independent), transcript on fs.blog", "type": "interview", "date": "2026-09-10", "publisher": "The Knowledge Project (podcast by Shane Parrish of Farnam Street, independent), transcript on fs.blog", "url": "https://fs.blog/knowledge-project-podcast/tobi-lutke-3/" }, "grade": "C", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "coding-agents", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/shopify-you-just-tell-ai-go-nuts-makes-2026", "json": "https://the-ai-record.com/api/v1/passages/shopify-you-just-tell-ai-go-nuts-makes-2026.json" }, { "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.", "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.", "speaker": "Mike Krieger", "speakerId": "mike-krieger", "speakerUrl": "https://the-ai-record.com/people/mike-krieger", "speakerLevel": "leader", "role": "Head of Anthropic Labs", "voice": "person", "organization": "Anthropic", "organizationId": "anthropic", "organizationUrl": "https://the-ai-record.com/organizations/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", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/anthropic-labs-talk-claude-about-code-instead-reading-every-line-2026", "json": "https://the-ai-record.com/api/v1/passages/anthropic-labs-talk-claude-about-code-instead-reading-every-line-2026.json" }, { "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.", "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.", "speaker": "Mike Krieger", "speakerId": "mike-krieger", "speakerUrl": "https://the-ai-record.com/people/mike-krieger", "speakerLevel": "leader", "role": "Head of Anthropic Labs", "voice": "person", "organization": "Anthropic", "organizationId": "anthropic", "organizationUrl": "https://the-ai-record.com/organizations/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", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-10-31" }, "topics": [ "code-review", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/anthropic-pull-requests-sent-with-artifact-explaining-intent-tradeoffs-2026", "json": "https://the-ai-record.com/api/v1/passages/anthropic-pull-requests-sent-with-artifact-explaining-intent-tradeoffs-2026.json" }, { "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.", "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.", "speaker": "Walden Yan", "speakerId": "walden-yan", "speakerUrl": "https://the-ai-record.com/people/walden-yan", "speakerLevel": "executive", "role": "Co-founder and Chief Product Officer", "voice": "person", "organization": "Cognition", "organizationId": "cognition", "organizationUrl": "https://the-ai-record.com/organizations/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", "gradeMeaning": "An interview or press quote.", "verification": { "status": "verified", "meaning": "An independent second reader re-fetched the source and confirmed the wording.", "reviewedOn": "2026-11-01" }, "topics": [ "code-review", "autonomy-and-oversight", "reading-agent-code", "failures-and-reversals" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/cognition-vibe-coding-no-review-experiment-two-weeks-2026", "json": "https://the-ai-record.com/api/v1/passages/cognition-vibe-coding-no-review-experiment-two-weeks-2026.json" }, { "id": "amazon-exceeded-max-rework-count-escalation-human-2026", "text": "And then you can see a whole bunch of these are merged and then we have some failures that say, hey, we've exceeded max rework count. And that's essentially an escalation to me saying, hey, you should probably look at this yourself as a human because you're responsible for the code that it generates.", "context": "Vaidya, an AWS solutions architect, is describing and demoing a dark factory (software built and merged by agents) that he built himself, which he calls \"my factory\"; it is not presented as an AWS product. The talk is from the Background Agents Summit, recorded 7 May 2026 and published by Ona.", "speaker": "Shardul Vaidya", "speakerId": "shardul-vaidya", "speakerUrl": "https://the-ai-record.com/people/shardul-vaidya", "speakerLevel": "practitioner", "role": "Solutions Architect, Data and AI, AWS", "voice": "person", "organization": "Amazon", "organizationId": "amazon", "organizationUrl": "https://the-ai-record.com/organizations/amazon", "aiVendor": true, "work": { "title": "Dark Factories: Verification-First Agentic Engineering", "type": "talk", "date": "2026-05-07", "publisher": "Background Agents Summit (YouTube)", "locator": "12:55", "url": "https://www.youtube.com/watch?v=e7AvdrxsbaU" }, "grade": "B", "gradeMeaning": "The organization's own post, paper or talk.", "verification": { "status": "reported", "meaning": "Not yet confirmed against the source by an independent second reader." }, "topics": [ "autonomy-and-oversight", "reading-agent-code" ], "functions": [ "engineering" ], "figures": [], "url": "https://the-ai-record.com/passages/amazon-exceeded-max-rework-count-escalation-human-2026", "json": "https://the-ai-record.com/api/v1/passages/amazon-exceeded-max-rework-count-escalation-human-2026.json" } ], "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 tools or services and have a reason to publish flattering numbers.", "An empty result means the archive holds nothing on it, not that nobody has said anything." ] }