DEV Community

Cover image for The Ex-GitHub CEO Just Launched a GitHub Rival — and the Pull Request Is the First Thing to Go
Paul Lovelace
Paul Lovelace

Posted on

The Ex-GitHub CEO Just Launched a GitHub Rival — and the Pull Request Is the First Thing to Go

#webdev #ai #github #programming

Disclosure: no affiliate links in this article. Everything below comes from Entire's October 26 launch announcement and the first coverage of it (GeekWire, MoneyControl, TechTimes). I have not used Entire hands-on; this is researched analysis, not a benchmark I ran, and Entire's speed and token claims are Entire's claims until independent testing checks them.

Thomas Dohmke ran GitHub. On October 26, 2026, he launched the thing meant to replace it.

His startup, Entire, unveiled what it calls the first Git forge built for humans and AI agents to work together. Hosting, review, builds — the core of what GitHub does — rebuilt for teams whose agents write code faster than people can review it. And the timing is not subtle: GitHub's annual Universe conference starts October 28.

If you have spent the last year drowning in AI-generated pull requests, this launch is aimed directly at you.

The pull request has become the bottleneck

Dohmke's argument is simple, and if you review agent code, you have felt it. The pull request was designed for humans proposing changes to other humans. Now agents propose most of the changes, and the pull request, in his words, "has become an audit log that only the agent reads."

Entire's replacement is called Trails. Instead of reading every changed line, you get a dashboard — Dohmke compares it to an airline pilot's heads-up display — with scores from AI agents evaluating the code on every push. Custom evals, scored again each time new code lands. Agents fix the findings in a loop. When every eval passes, you merge, without necessarily reading the code yourself.

You can still require a human sign-off, or passing automated tests, before anything merges. That matters. "Merge without reading the code" is a sentence that should make any experienced developer nervous, and Entire knows it — the human gate is optional, not absent. My take: use it. At least until you have watched Trails catch real bugs on your own codebase for a few weeks.

Trails data is available on Entire's web interface and through its open-source CLI, so agents can work the loop themselves while you watch the scores move.

The other two pieces: faster Git, and memory that survives the session

Distributed Git hosting. Entire built a new source-control server compatible with Git. It claims pushes up to 89 times faster than GitHub and 6 times faster than Cursor Origin, and calls itself the fastest Git forge in production today. Repositories can be hosted in the region you choose — US, EU, India, and Australia at launch.

That claim lands differently when you remember GitHub's recent year: several significant outages, including one in August that lasted nearly eight hours, and a migration to Microsoft's Azure cloud to add capacity. GeekWire notes developers and agents made 7.38 billion commits on GitHub in September alone. When agents push constantly, hosting speed and uptime stop being nice-to-haves. Treat the 89x figure as a vendor claim — your repo, your network, your mileage — but the direction is right: agent-era Git needs to be fast and distributed, not a single queue everyone waits in.

Continuous Memory. This is the piece I would actually try first. Entire's entire-brain tool stores agent session histories in Git: the idea, the agent trace, and the outcome, as first-class artifacts alongside the code. When an agent meets the same code again, it can surface why a decision was made last time instead of re-deriving it from scratch. Entire claims this cuts token usage by up to 71%.

That number is Entire's, not mine, and token savings depend entirely on how repetitive your sessions are. But the idea is sound and it builds on something real: Entire's first product, Checkpoints, launched in February, was an open-source tool that captured an agent's prompts, decisions, and session history alongside code changes. Continuous Memory is that idea grown up — memory as part of the repo, not locked inside one agent's chat history.

Entire also announced CI integrations with Buildkite and Depot, and says the platform is available to every developer today, backed by $60 million in funding.

Should you switch? Not this week. Should you try it? Yes — on a side repo

Here is the honest version, because a launch announcement is not a migration plan:

Try Trails on a real but non-critical repo. Pick something with a decent test suite, turn on the human sign-off gate, and watch what the evals catch and what they miss for two weeks. The question is not whether agent-scored review works in a demo. It is whether it catches the bug your senior reviewer would have caught.

Do not migrate your main repo on launch week. Entire is new. Its uptime, its support, its pricing beyond the announcement — all unproven at scale. GitHub, for all its outages, is the place your contributors already are. A second forge adds friction: contributors, CI, permissions, and institutional memory all have to move too.

Steal the idea even if you stay on GitHub. The real insight here is not "leave GitHub." It is that review has to change when agents write most of the code. If you stay, ask yourself: what are your evals? What gets checked on every push, automatically, with a score you can read at a glance — not a wall of diff? If the answer is "a human reading everything," you have the bottleneck Dohmke is selling against, whether or not you buy Entire.

Watch Universe. GitHub's conference starts October 28, two days after this launch. If GitHub answers Trails with its own agent-review story, that tells you this is the new battleground, not a one-company quirk. Either way, the pull request as we knew it is on borrowed time.

What to do this week

  1. Count your review bottleneck. How many open pull requests are waiting on a human right now, and how many of them were written mostly by an agent? That number is your case for trying something like Trails.
  2. Try Entire on one side project. Turn on human sign-off. Keep your existing tests as a required gate. Compare what Trails flags against what your own review finds.
  3. Audit your agent memory. When an agent revisits old code in your team, does it know why a decision was made — or does it guess again? If it guesses, a memory layer (Entire's or anyone's) is worth testing.
  4. Do not let any tool merge to main unsupervised yet. Agent-scored review, human-scored review, it does not matter. New review tooling earns trust on side branches first.

The ex-CEO of GitHub betting against the pull request is not a small story. It is a sign that the workflow every developer learned in the last fifteen years is being rewritten for the agent era — and the rewrite starts with the part we all hate most: the review queue.

FAQ

What is Entire?
A developer platform launched October 26, 2026, by former GitHub CEO Thomas Dohmke — a Git forge (code hosting, review, and builds) built for humans and AI coding agents to work together, positioned as an alternative to GitHub.

What are Trails?
Entire's replacement for pull requests: agent-scored evals on every code push, shown as a heads-up dashboard. Agents fix findings in a loop, and you merge once evals pass — with an optional human sign-off or passing-tests gate.

How much faster is Entire's Git hosting?
Entire claims pushes up to 89 times faster than GitHub and 6 times faster than Cursor Origin, on a new Git-compatible server with hosting regions in the US, EU, India, and Australia. That is a vendor claim, not an independent benchmark.

What is Continuous Memory?
Entire's entire-brain tool stores agent session histories — ideas, traces, and outcomes — in Git, so future agent sessions can reuse past decisions. Entire claims up to 71% lower token usage; that figure is vendor-reported.

Is Entire available now?
According to its launch announcement, yes — available to every developer today, with CI integrations for Buildkite and Depot.

Sources

Top comments (0)