Back to News
IndustryAI Understanding briefing

Former GitHub CEO Thomas Dohmke launches Entire, an agent-native Git platform

Entire, a new Git forge founded by former GitHub CEO Thomas Dohmke, has launched with a focus on agent-native workflows, automated code evaluation, and persistent memory for AI coding agents.

4 min readRead the linked source
Source-provided image accompanying Former GitHub CEO Thomas Dohmke launches Entire, an agent-native Git platform
Source referenceSource recorded
Publisher
itwire.com
Source type
Linked source — primary-source status has not been established.
Also cited

Story last revised

ContextUnderstand this in 60 seconds

Key terms

Memory (Agent Memory)
Stored context an AI agent uses across steps or sessions to improve continuity.
Token
A chunk of text processed by language models, such as a word piece or symbol.
Test yourselfAI Agents Quiz

What changed since publication

  1. First published
  2. The platform has officially launched, providing further details on its 'Trails' review system, 'Continuous Memory' architecture, and regional hosting capabilities, following its initial announcement.

What happened

Former GitHub CEO Thomas Dohmke has launched Entire, a new Git-compatible developer platform designed specifically for AI coding agents. The platform introduces 'Trails,' a review system that prioritizes agent-generated evaluations over traditional code diffs, and 'Continuous Memory,' which stores agent session history to improve context retention. Entire also offers regional repository hosting, including in Australia, and claims significantly higher push throughput compared to GitHub and Cursor Origin.

Entire, founded by former GitHub CEO Thomas Dohmke, has officially launched its Git-compatible platform. The service is built to accommodate the high volume of code generated by AI agents, replacing traditional pull requests with 'Trails,' a system that organizes reviews around automated evaluations rather than just code diffs and comments.

The platform features 'Continuous Memory,' which stores agent prompts and session history alongside repository data. The company claims this graph-based approach can reduce consumption by up to 71% by allowing agents to retrieve knowledge from previous work rather than re-processing entire codebases.

Entire provides native private repository hosting in Australia, the US, EU, and India. It includes integrations with Buildkite and Depot for continuous integration, with migration tools available for teams moving from GitHub Actions.

The company claims its platform supports up to 418 pushes per second, citing internal benchmarks that suggest it is significantly faster than GitHub and Cursor Origin. However, the source notes these figures have not been independently verified and may be influenced by network distance and GitHub's own rate-limiting policies.

Source details: itwire.com

Why it matters

As AI coding agents generate code at speeds exceeding human review capacity, Entire attempts to solve the resulting bottleneck by shifting the review paradigm from manual line-by-line inspection to automated evaluation-based workflows. By treating agent traces and reasoning as first-class artifacts, the platform aims to maintain developer oversight while scaling production. The platform's success depends on its ability to integrate into existing enterprise CI/CD pipelines and provide the security and governance required for professional software development, while competing against incumbents like GitHub that are also integrating agent workflows.

The core challenge Entire addresses is the 'review bottleneck' created by AI agents. When agents generate code faster than humans can inspect it, traditional review processes become inefficient. Entire attempts to mitigate this by allowing developers to merge changes based on automated evaluation results, while still providing configurable 'merge gates' for human approval.

By treating the 'agent trace' and the 'idea' behind a code change as first-class artifacts, Entire aims to preserve the reasoning behind software changes, which is often lost in standard Git commits. This is intended to reduce redundant investigations and improve the long-term maintainability of AI-generated codebases.

The platform is backed by a $60 million seed round led by Felicis at a $300 million valuation, with notable participation from Microsoft’s M12 fund. Despite the backing, the company faces stiff competition from GitHub itself, which is actively developing its own agent-native workflows.

Interactive Mechanism

Interactive Mechanism: How It Actually Works

Explore the underlying technology behind this development interactively.

Thinking Budget (Test-Time Tokens):1,024 tokens
Complex Accuracy79%Math & Code Logic
Latency3.2sTime to first full output
Inference Cost$0.0092Per query estimated
Reasoning StyleStep VerificationInternal chain depth
Active Thinking Trace:
1Deconstruct user problem into formal constraints
2Propose candidate hypotheses & step-by-step calculation
3Self-correction: Backtrack and refute subtle edge cases
4Exhaustive consistency check & final output synthesis
Core takeaway: Test-time compute fundamentally changes AI economics. Instead of only scaling during pre-training, giving reasoning models more tokens at inference time allows them to systematically solve PhD-level STEM problems.
Interactive Concept Check+10 Points
AI Agents Quiz

An agent must create a draft calendar event for Tuesday at 2 p.m. Which evidence would establish the requested result?

What to watch next

Entire's roadmap includes the addition of public repositories, self-hosted Git nodes, and live-streamed sessions, which are currently unavailable. Potential users should monitor how the platform handles security and compliance, as the company notes that automated evaluations do not guarantee code quality or security. Additionally, the industry will observe whether the platform's performance claims hold up in real-world, high-volume production environments compared to established competitors.

The platform is currently limited by the absence of public repositories, self-hosted Git nodes, and live-streamed sessions, all of which are on the company's roadmap. These features are critical for broader adoption beyond early-adopter teams.

Security and governance remain significant unknowns. Entire explicitly warns that passing automated evaluations does not ensure software is free of security flaws or compliant with business requirements, meaning teams must still define their own accountability and review standards.

The company's performance claims regarding push throughput are based on internal benchmarks. Prospective users should evaluate these claims within their own network environments, as the source highlights that network distance and external rate limits are major variables in such performance metrics.

Related guides & quizzes

Updates and corrections

This canonical story is updated in place when the developing event materially changes. Its URL and original publication date never change.

  • The platform has officially launched, providing further details on its 'Trails' review system, 'Continuous Memory' architecture, and regional hosting capabilities, following its initial announcement.
See the public corrections log
Found this useful?