Former Thomas Dohmke CEO launches Git platform Entire to save us from AI code overload
While GitHub is busy drowning in Azure migrations and random outages, its former boss just built a custom playground for the AI hordes that are currently breaking the internet's coding infrastructure.
Thomas Dohmke, who stepped down as GitHub CEO to escape the corporate thrills of Microsoft's bureaucracy, officially launched Entire. The new platform is designed specifically to handle the relentless, high-frequency code commits generated by AI coding agents. Unlike human developers who need coffee breaks and sleep, AI bots tend to spam repositories with endless updates, leaving traditional hosting platforms gasping for air.
The platform allows developers to mirror their public and private GitHub repositories directly into Entire's regional mirrors. AI agents like Copilot, Claude Code, and Cursor can then clone, read, and rewrite code locally without triggering rate limits. It is essentially a giant shield protecting old-school servers from the digital equivalent of a locust swarm.
To help humans understand what their silicon assistants are actually doing, the platform tracks full AI agent sessions. It records every prompt, response, file edit, and reasoning step. The built-in "Blame" and "Review" features allow human supervisors—or other AI agents, because why not automate the bureaucracy too—to audit precisely who modified a line of code and what prompt triggered it.
Benchmark tests show the network handles 2.1 million push operations and 570,000 clones per hour. By comparison, rival agent-focused Git platform Cursor Origin maxes out at just 81,000 pushes and 296,000 clones. This massive capacity comes at a time when GitHub is frozen in a two-year infrastructure migration to Azure, prioritizing database moving over actual feature development.
The spectacle of a former chief building a life raft to rescue developers from his own previous company's crumbling infrastructure is peak tech industry. As human coders slowly transition into prompt-writing managers, the real battle is no longer about who writes the best code, but whose servers can survive the sheer volume of automated garbage being pushed to production.
Source: Entire
Comments
This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.