Former GitHub CEO Dohmke launches Entire platform built for AI coding agents

4 Sources

Share

Former GitHub CEO Thomas Dohmke unveiled Entire, a developer platform designed for AI coding agents that replaces pull requests with automated Trails evaluations. The platform offers Git hosting up to 89x faster than GitHub and includes Continuous Memory to reduce token usage by 71%.

Former GitHub CEO Unveils Agent-First Platform

Former GitHub CEO Thomas Dohmke launched Entire on October 26, 2026, positioning the Git-compatible development platform as a GitHub rival designed specifically for AI coding agents

1

2

. After nearly four years leading GitHub, Dohmke left in 2025 to build a solution addressing a critical bottleneck: AI-generated code now arrives faster than developers can review it

2

. The timing proves strategic, launching just two days before GitHub's annual Universe conference in San Francisco and as GitHub struggles with mounting outages from AI coding activity

2

.

Source: GeekWire

Source: GeekWire

Trails Replace Traditional Pull Requests

Entire's signature feature, Trails, fundamentally reimagines code review for the AI era. Instead of traditional pull requests where developers manually examine every changed line, Trails provides a dashboard showing automated evaluations scored by AI coding agents on every code push

1

. "Unlike the classic Pull Request model of reviewing code changes with a diff, Trails offer developers a heads-up display similar to an airline pilot, with customized evals that are scored by agents," the company explained

1

. Multiple AI coding agents assess code quality, security, and other configurable criteria, allowing developers to merge code confidently without reviewing every line

4

. Teams retain control by setting requirements like human approvals or passing automated tests before merging

3

.

Source: ZDNet

Source: ZDNet

Addressing GitHub's Infrastructure Strain

The agent-first platform arrives as GitHub faces significant infrastructure challenges. Developers and AI coding agents made 7.38 billion commits on GitHub in September 2026 alone, contributing to several major outages including an eight-hour disruption in August

2

. GitHub is migrating from its own data centers to Microsoft's Azure cloud to add capacity

2

. Entire's decentralized Git network, announced in July, specifically targets this problem. The platform claims Git hosting up to 89x faster than GitHub and 6x faster than Cursor Origin based on open-source benchmarks

1

3

.

Continuous Memory Reduces Token Consumption

Continuous Memory represents Entire's solution to context management for AI coding agents. The feature keeps developer session logs, prompts, and decisions alongside code in repositories, creating searchable data sources that help agents avoid repeating mistakes

1

. This approach mirrors memory features in AI chatbots that use past conversations to speed responses. Entire reports Continuous Memory reduced token usage by up to 71% in testing through its codebase graph and "agentic brain" that learns from previous sessions

1

3

. This matters as token costs accumulate rapidly when AI coding agents repeatedly gather context.

Global Data Hosting Including India

The developer platform enables private repository hosting across multiple regions: the US, India, Australia, and the EU

1

3

. India represents a key market, where GitHub reported 27 million developers as of April 2026, adding over five million during 2025 according to its Octoverse report

3

. Entire appointed former GitHub APAC Field CTO Karthik Rameshkumar as Field CTO for India in July to drive adoption

3

. The distributed hosting helps organizations meet data residency requirements while speeding code access for AI coding agents

3

.

Competing in Crowded Developer Platform Market

Entire enters competitive territory with its 50-person fully remote team

2

. The company raised a $60 million seed round in February led by Felicis, with participation from Microsoft's M12 venture arm, Seattle-based Madrona, Basis Set Ventures, and individual investors including Yahoo co-founder Jerry Yang and Y Combinator CEO Garry Tan

2

. Other platforms are also adapting to AI-driven code reviews: Cursor launched its Origin code-hosting service in August, while Anthropic built its own source control system as daily code merges jumped from 2,500 in May to over 30,000 in September

2

. Entire partners with Buildkite and Depot for build and test pipelines, offering an alternative to GitHub Actions

2

.

Why This Matters for Software Development

Dohmke frames the shift bluntly: "Code review was always a bottleneck. 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," he said at a Madrona event

2

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

2

. However, automated evaluations don't eliminate the need for human judgment on security, correctness, and critical software decisions

4

. Entire's success depends on whether development teams trust AI-driven code reviews enough to adopt the platform while maintaining appropriate oversight. Developers can mirror existing GitHub repositories or migrate entirely, providing flexibility to test the approach

2

3

. Watch for adoption metrics among enterprise teams and how Entire handles edge cases where automated evaluations miss critical issues.

© 2026 TheOutpost.AI All rights reserved