Former GitHub CEO Launches Entire, a GitHub Rival Built for AI Coding Agents

Former GitHub CEO Launches Agent-First Platform

Former GitHub CEO Thomas Dohmke has officially launched Entire, a GitHub rival built specifically for AI coding agents. The developer platform, now available for use, addresses the mounting pressure on traditional coding infrastructure as AI-generated code accelerates development cycles.1

1

2

2

Dohmke, who left GitHub last year after nearly four years as CEO, secured $60 million in seed funding led by Felicis, with participation from Microsoft's M12 venture arm, Seattle-based Madrona, and Basis Set Ventures.2

2

The timing is strategic. GitHub has experienced several significant outages in recent months, including an eight-hour disruption in August, as the platform strains under increased load from AI coding agents. Developers and agents made 7.38 billion commits on GitHub in September alone, forcing the platform to migrate from its own data centers to Microsoft's Azure cloud to add capacity.2

2

Entire now employs 50 people and operates as a fully remote company.

Trails Replace Traditional Pull Requests

The most significant innovation Entire introduces is Trails, a replacement for the traditional pull request model. Instead of reviewing code changes line by line through diffs, Trails provide developers with a heads-up display similar to an airline pilot's interface, featuring customized evaluations scored by AI coding agents on every code push.1

1

Multiple AI agents check the quality of each change, allowing developers to merge code without reading every line.

"With Trails, Entire treats the idea, the agent trace, and the outcome of a developer's sessions as first-class artifacts for review alongside the code," Dohmke explained.1

1

The system uses multiple agents to generate evals that verify the quality and success of AI-generated code, enabling developers to confidently merge changes without manual review of every line in the diff. Teams can still require human sign-off or passing automated tests before code merges.2

2

The agents running this process are customizable, and each trail displays how agents are spending tokens, providing transparency into the evaluation process.1

1

All data is accessible through Entire's web interface and the open-source Entire CLI, allowing coding agents to address findings in a loop while human developers validate progress.

Decentralized Git Network Delivers Speed

Entire claims to offer Git hosting up to 89x faster than GitHub and six times faster than Cursor Origin, thanks to its decentralized Git network announced in July.1

1

2

2

The platform lets developers create, migrate, and host private repositories in multiple locations across the world, currently starting with the U.S., India, Australia, and the EU.1

1

This decentralized approach directly addresses scalability issues plaguing GitHub. The network is designed to withstand the sharp increase in AI coding activity that has caused mounting GitHub outages.1

1

Entire is partnering with Buildkite and Depot on build and test pipelines, giving developers a way to move off GitHub Actions, while public repositories and self-hosting remain on the roadmap.2

2

Continuous Memory Reduces Token Usage

Another differentiating factor is Continuous Memory, which keeps developer session logs next to code in the repository as a data source. AI coding agents can search these logs for answers, helping them use up to 71% fewer tokens according to Entire's testing.1

1

2

2

This feature is similar to memory capabilities in AI chatbots that reduce time spent gathering context by using past conversations or other agent interactions.

Continuous Memory comprises a codebase graph and an "agentic brain" that uses previous sessions and prompts as lessons so agents don't repeat mistakes.1

1

This approach lets AI coding agents search past coding sessions for the context behind earlier changes, streamlining the development process and reducing token usage costs.2

2

Addressing the Code Review Bottleneck

"Code review was always a bottleneck," Dohmke said at a Madrona event last month. "I've never worked in a software company where I didn't wait for somebody else to review my pull request, and you have to ping people in Slack three times. And there was this one guy that had to review everything because he knew what to look for. And now that bottleneck is 10 times bigger."2

2

The pull request "has become an audit log that only the agent reads," Dohmke wrote in a blog post.2

2

This observation drives Entire's fundamental redesign of the code review process for an era where AI coding agents write code faster than people can review it.

Entire isn't alone in rethinking code hosting for AI coding agents. AI coding company Cursor launched its own code-hosting service, Origin, in August. Anthropic recently described building its own source control system as the number of code changes it merged each day grew from about 2,500 in May to more than 30,000 in September.2

2

Watch how established players like GitHub respond to this agent-first platform challenge and whether Entire's approach to handling AI-generated code becomes the industry standard.