--- title: Ex-GitHub CEO bets the pull request is dead | daily.dev description: Former GitHub CEO Thomas Dohmke launched Entire, a Git forge built for agent-human collaboration rather than traditional pull requests. Its core feature,... canonical: https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Ex-GitHub CEO bets the pull request is dead | daily.dev og:description: Former GitHub CEO Thomas Dohmke launched Entire, a Git forge built for agent-human collaboration rather than traditional pull requests. Its core feature,... og:url: https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p og:image: https://api.daily.dev/og/posts/rCjBzCW6p.png og:image:alt: Ex-GitHub CEO bets the pull request is dead og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Ex-GitHub CEO bets the pull request is dead **[Trends](https://daily.dev/sources/trends)** · 2 min read · 32 upvotes · 4 comments ## Summary Former GitHub CEO Thomas Dohmke launched Entire, a Git forge built for agent-human collaboration rather than traditional pull requests. Its core feature, Trails, replaces diffs and comment threads with a dashboard of customizable evals that run on every push, scoring code for security, risk, intent drift, and 'slop'. Entire claims its distributed server pushes code up to 89x faster than GitHub and 6x faster than Cursor's Origin, though these are vendor benchmarks. The platform is free, charging only for tokens used by review agents, with new users getting a $100 credit. It also ships open-source tools entire-graph and entire-brain, and launches CI integrations with Buildkite and Depot. ## Content The sharpest framing so far comes from @GergelyOrosz, who says Entire is "the ONLY company I've invested in the last two years." His reasoning: "we need an 'AI-native' GitHub, and also clear that GitHub cannot disrupt itself." That's an investor putting his name behind the idea that the incumbent is stuck. The incumbent in question used to be run by the guy building the rival. Thomas Dohmke, former GitHub CEO, just launched Entire, a Git forge pitched at agent-human collaboration. @ashtom (the handle posting the launch thread) frames it bluntly: "Agents reinvented how we produce code. It changed our tools and half of our workflow. It's time for the other half to catch up." The team says it built Entire using Entire. ## The pull request gets replaced The headline feature is Trails, which swaps the pull request for a dashboard of customizable evals that agents score on every push. They cover things like security, risk, intent drift and "slop" detection. You can run them from the web interface or an open-source CLI. That is the real swing here. Not a nicer PR page, but the claim that review should be automated scoring tuned for agent-sized volumes of code. ## The speed claims Entire says its Git-compatible server, backed by a new backend called EntireDB, pushes up to 89x faster than GitHub and 6x faster than Cursor's Origin. Those are the company's own numbers, so treat them as a marketing slide until someone independent reproduces them. Private repos are hosted in the US, Europe, Australia and India. ## Memory and pricing Continuous Memory syncs agent sessions into the repo through two open-source CLI tools, entire-graph and entire-brain. They index code structure and agent sessions. In Entire's own testing, entire-graph saved up to 71% on tokens. CI integrations start with Buildkite and Depot. The platform is free. You pay only for the tokens the review agents consume, and you start with a $100 credit. ## Where this leaves things The pitch is clean: GitHub was built for humans pushing a few commits, and agents don't work like that. Whether a startup can pull teams off the place where all their code, issues and CI already live is the open question. I'd watch the first independent push benchmarks, and whether anyone trusts eval scores enough to merge on them. ## Questions this post answers ### What is Entire, the new Git platform from the former GitHub CEO? Entire is a Git forge built by Thomas Dohmke, former GitHub CEO, designed for agent-human collaboration instead of traditional pull requests. Its core feature, Trails, replaces diffs and comment threads with a dashboard of customizable evals that run on every push, scoring code for security, risk, intent drift, and 'slop'. It also ships open-source CLI tools entire-graph and entire-brain. _Teams weighing AI-native code review platforms can track launches like this one on daily.dev._ ### How does Entire's pricing model work compared to GitHub? Entire is free to use, but you pay for the tokens consumed by review agents rather than per-seat licensing, with new users starting with a $100 credit. This ties cost directly to agent compute usage rather than headcount, a different model from GitHub's seat-based pricing. _Developers comparing AI coding tool pricing models can follow cost-structure shifts like this on daily.dev._ ### How much faster does Entire claim its Git server is than GitHub? Entire claims its Git-compatible distributed server delivers pushes up to 89 times faster than GitHub and 6 times faster than Cursor's Origin. These figures come from Entire's own benchmarks and have not been independently verified, so they deserve skepticism until outside testing confirms them. _Anyone deciding between Git hosting options can watch for independent benchmarks via daily.dev._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 46 comments across hackernews (as of 2026-10-27). **TL;DR:** Reaction is split between curiosity about the session/checkpoint and graph-indexing ideas and skepticism about whether reviewing agent sessions is actually valuable, with several pointed questions about the product's premise, pricing, and marketing tone. **Sentiment:** 25% positive · 30% mixed · 45% skeptical **The case for** - A few users report having used the checkpoints and Trails features for weeks and say they wouldn't go back. - The code-graph generator is seen as a genuinely interesting piece of the release. **The pushback** - Some question whether attaching full agent session context to commits adds real debugging value versus just being more to review. - Doubt that storing proprietary agent session data with a third party is worth it when context can be stored elsewhere. - The blog post and 'rebellion' framing are criticized as overly marketing-heavy for a venture-backed startup. - Skepticism that reviewing 'Trails' instead of diffs is actually where PR review should be heading. - The core premise that humans will stop writing most code is challenged as a cynical bet rather than a real belief about tooling's future. **By community** - hackernews (mixed): A company representative actively engaged with pointed questions about session value, data ownership, pricing, and the startup's core premise, producing a debate that mixed curiosity, pushback, and a few enthusiastic user endorsements. **Hottest debate:** Whether attaching full agent session context to commits provides real reviewable value or is just unnecessary overhead compared to the diff itself. **Open questions** - What exactly is lost between a session transcript and the resulting code that would actually help debug something? - If agents write most of the code, why would human PR review still be the right model at all? - Is the main value proposition just time savings for large dev orgs, and is that enough to justify handing over proprietary agent context? **Highlights** > Would there have been value, pre-LLMs, in attaching a screen recording of what a developer did in the three hours before posting their diff (be it Google, stack overflow, reading a book, or trying a sandbox experiment) to their PR? The PR is a synthesis of their efforts, and it would actually be more effort to review and understand their screen recordings, wouldn't it? In my experience striving for fewer artifacts is the right move, so long as they are sufficient to describe the outcome. What exactly is lost between sessions and code that would help you debug something? Also, if your business has agents writing most of the code, why would your business care about human reviews at all? In for a penny, in for a pound, after all. Wouldn't you be better off looping your agents into your support tickets and autofixing bugs? In that scenario, isn't it more confusing for an agent to deal with sessions + code rather than start fresh with the code? > — [argee on hackernews · 1 comments](https://news.ycombinator.com/item?id=50248036) > "We founded Entire on a single belief, human developers no longer write most of the code." This isn't a belief. What do you actually believe about the future of software? Let me put it this way. The premise of agents is that some combination of programming languages and programming tools suck so incredibly much that humans don't want to use them. You're effectively betting that this problem can never be fixed. AI is a layer of kludges on top of the problem of bad programming tools. Haven't you ultimately made a rather cynical bet? > — [conartist6 on hackernews · 1 comments](https://news.ycombinator.com/item?id=50248113) > The graph generator looks interesting, but the rest of the blog post is filled with so much marketing it's a slog to get through to the good bits. I struggle to imagine a scenario where I willingly choose the "Trails" view even if, and god help us if it is, that's the direction PR review is going. > — [stryan on hackernews · 2 comments](https://news.ycombinator.com/item?id=50247949) > i guess the main thing is "why would give you my proprietary agent-facing information when i can store context literally anywhere?" is the carrot JUST the time savings for large dev orgs? > — [0gs on hackernews · 1 comments](https://news.ycombinator.com/item?id=50248068) > > Join the rebellion This feels insincere coming from a venture-backed software startup. Do you have a unique ownership structure, or a particular emphasis on open-source or self-hosting? > — [salt-licker on hackernews · 1 comments](https://news.ycombinator.com/item?id=50248239) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=50247453) · 42 points · 46 comments ## Community discussion Top comments from developers on daily.dev. **@parsam** · 15 upvotes > the diff review is definitely burning people out when agents generate ten files at a time. but the real bottleneck is we're still reading line by line for system-level failures like pooled connections not returned on an error path, which a diff view hides because the setup was in another file three commits ago. **@pdfopsdev** · 4 upvotes > scoring every push for 'slop' sounds neat until the eval model itself drifts and now you're debugging the evaluator instead of the diff. does Trails let you pin eval versions, or are you re-baselining every time the underlying LLM updates? **@diyweb** · 3 upvotes > The point about reviewing line by line for system-level failures is the real test for Trails. A score is only useful if it links back to the decision that caused it, so a reviewer can see why something was flagged and not just that it was. Without that context it becomes another dashboard people learn to ignore. **@skepagent** · 2 upvotes > 「评审瓶颈先于生成瓶颈崩掉」这个判断我挺认同，智能体一次吐十几个文件时，逐行看 diff 确实跟不上。Trails 如果真能把上下文和决策理由留在可追溯的位置，而不是只给一个打分面板，那它才算解决了问题，否则换个界面还是同一份负担。 ## Similar posts on daily.dev - [GitHub’s former CEO launches a developer platform for the age of agentic coding](https://daily.dev/posts/github-s-former-ceo-launches-a-developer-platform-for-the-age-of-agentic-coding-dphwu4bkt) · The New Stack · 1 upvotes · 0 comments - [Former GitHub CEO raises record $60M dev tool seed round at $300M valuation](https://daily.dev/posts/former-github-ceo-raises-record-60m-dev-tool-seed-round-at-300m-valuation-5tblooavz) · TechCrunch · 1 upvotes · 0 comments - [Hello Entire World · Entire](https://daily.dev/posts/hello-entire-world-entire-t3wpt9uvy) · Hacker News · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#github](https://daily.dev/tags/github), [#git](https://daily.dev/tags/git), [#code-review](https://daily.dev/tags/code-review) [View this post on daily.dev](https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Ex-GitHub CEO bets the pull request is dead","url":"https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p"},"datePublished":"2026-10-26T19:13:08.263Z","dateModified":"2026-10-27T17:03:07.975Z","description":"Former GitHub CEO Thomas Dohmke launched Entire, a Git forge built for agent-human collaboration rather than traditional pull requests. Its core feature,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a81bff9422e9ac5eb43865f382792124?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a81bff9422e9ac5eb43865f382792124?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":4,"discussionUrl":"https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":32},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":4}],"keywords":"ai-agents,github,git,code-review","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Ex-GitHub CEO bets the pull request is dead"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p","comment":[{"@type":"Comment","text":"the diff review is definitely burning people out when agents generate ten files at a time. but the real bottleneck is we’re still reading line by line for system-level failures like pooled connections not returned on an error path, which a diff view hides because the setup was in another file three commits ago.","datePublished":"2026-10-26T19:23:20.043Z","url":"https://daily.dev/posts/rCjBzCW6p#c-aCjIlQYfq","author":{"@type":"Person","name":"Parsa","url":"https://daily.dev/parsam","image":"https://lh3.googleusercontent.com/a/ACg8ocI7cLcjVdmLqIn3k8yJcwf5xcUYqgUd6d2dh6oiY59utH1geA=s96-c"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":15}},{"@type":"Comment","text":"scoring every push for ‘slop’ sounds neat until the eval model itself drifts and now you’re debugging the evaluator instead of the diff. does Trails let you pin eval versions, or are you re-baselining every time the underlying LLM updates?","datePublished":"2026-10-27T18:01:23.503Z","url":"https://daily.dev/posts/rCjBzCW6p#c-emHdDjba3","author":{"@type":"Person","name":"PDFops","url":"https://daily.dev/pdfopsdev","image":"https://media.daily.dev/image/upload/s---8isRBKc--/f_auto/v1782922291/avatars/avatar_orjMeK8QKaaVZwGq7ScPz?_a=BAMAMicg0"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":4}},{"@type":"Comment","text":"The point about reviewing line by line for system-level failures is the real test for Trails. A score is only useful if it links back to the decision that caused it, so a reviewer can see why something was flagged and not just that it was. Without that context it becomes another dashboard people learn to ignore.","datePublished":"2026-10-27T18:02:15.820Z","url":"https://daily.dev/posts/rCjBzCW6p#c-cWKkwTUiD","author":{"@type":"Person","name":"DiyWeb","url":"https://daily.dev/diyweb","image":"https://media.daily.dev/image/upload/s--dgS6jDeN--/f_auto/v1790879420/avatars/avatar_774zDwMaRTNS1fv5OKStX?_a=BAMAMicg0"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":3}},{"@type":"Comment","text":"「评审瓶颈先于生成瓶颈崩掉」这个判断我挺认同，智能体一次吐十几个文件时，逐行看 diff 确实跟不上。Trails 如果真能把上下文和决策理由留在可追溯的位置，而不是只给一个打分面板，那它才算解决了问题，否则换个界面还是同一份负担。","datePublished":"2026-10-27T03:09:17.414Z","url":"https://daily.dev/posts/rCjBzCW6p#c-RGG2WEsGD","author":{"@type":"Person","name":"kris gu","url":"https://daily.dev/skepagent","image":"https://lh3.googleusercontent.com/a/ACg8ocJnnwFKNx7PJNhb7W6mFWr2k2SV_VHW_9ainY71DX_dv79frQ=s96-c"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ex-github-ceo-bets-the-pull-request-is-dead-rcjbzcw6p#faq","mainEntity":[{"@type":"Question","name":"What is Entire, the new Git platform from the former GitHub CEO?","acceptedAnswer":{"@type":"Answer","text":"Entire is a Git forge built by Thomas Dohmke, former GitHub CEO, designed for agent-human collaboration instead of traditional pull requests. Its core feature, Trails, replaces diffs and comment threads with a dashboard of customizable evals that run on every push, scoring code for security, risk, intent drift, and 'slop'. It also ships open-source CLI tools entire-graph and entire-brain. Teams weighing AI-native code review platforms can track launches like this one on daily.dev."}},{"@type":"Question","name":"How does Entire's pricing model work compared to GitHub?","acceptedAnswer":{"@type":"Answer","text":"Entire is free to use, but you pay for the tokens consumed by review agents rather than per-seat licensing, with new users starting with a $100 credit. This ties cost directly to agent compute usage rather than headcount, a different model from GitHub's seat-based pricing. Developers comparing AI coding tool pricing models can follow cost-structure shifts like this on daily.dev."}},{"@type":"Question","name":"How much faster does Entire claim its Git server is than GitHub?","acceptedAnswer":{"@type":"Answer","text":"Entire claims its Git-compatible distributed server delivers pushes up to 89 times faster than GitHub and 6 times faster than Cursor's Origin. These figures come from Entire's own benchmarks and have not been independently verified, so they deserve skepticism until outside testing confirms them. Anyone deciding between Git hosting options can watch for independent benchmarks via daily.dev."}}]} ```