San Francisco: Former GitHub chief executive Thomas Dohmke has launched Entire, a developer platform designed to help human programmers and artificial intelligence (AI) coding agents work together. The platform aims to challenge GitHub by rethinking how software code is hosted, reviewed, tested and delivered as AI tools increasingly generate large volumes of code.
Unveiled on October 26, Entire introduces an alternative to traditional software development workflows, with features designed to help teams review AI-generated changes more efficiently. Its key offerings include Trails, a new approach to code reviews, distributed Git hosting and Continuous Memory, which allows AI agents to draw on context from previous coding sessions.
The launch comes as software developers increasingly rely on AI agents to write code, fix bugs and perform technical tasks. As these tools accelerate code production, reviewing and validating their output has become a growing challenge for development teams.
Why Thomas Dohmke is challenging GitHub
Dohmke, who left GitHub in 2025 after nearly four years as its chief executive, founded Entire to develop a platform built specifically for collaboration between people and AI agents.
GitHub, owned by Microsoft, has long been a central platform for software developers to store code, collaborate on projects and review changes before they are incorporated into a software product.
However, traditional development workflows were designed around people writing code and submitting changes for other developers to inspect. AI coding agents can generate changes much faster, potentially creating more code for teams to review than human reviewers can comfortably handle.
Entire argues that software development platforms must evolve to accommodate this shift. Rather than simply adding AI assistants to existing systems, the company is redesigning parts of the development process around agents that can write, evaluate and improve code.
The platform is positioned as a direct alternative to GitHub, although Entire’s longer-term ambition extends beyond code hosting to managing the wider software development lifecycle.
Trails replaces the traditional pull-request approach
One of Entire’s main features is Trails, which offers an alternative to the conventional pull-request process used by GitHub and other code-hosting platforms.
In a traditional workflow, a developer submits a pull request containing proposed code changes. Other developers then review the differences, identify possible problems and approve or reject the changes before they are merged into the main project.
Entire argues that this process can become a bottleneck when AI agents generate code at high speed.
Trails instead uses multiple AI agents to evaluate code changes whenever new code is submitted. The system presents developers with assessments and scores covering the quality of the changes, helping teams identify potential problems without manually inspecting every line first.
AI agents can then address issues identified during the review process, while human developers monitor progress and decide whether the changes are ready to be merged.
Entire says teams can establish conditions that must be satisfied before code is accepted. These can include successful automated tests or explicit human approval.
The approach is intended to make code review more scalable without eliminating human oversight. Developers remain responsible for setting requirements, checking results and deciding when changes can enter a project.
Entire claims Git hosting can be up to 89 times faster
Another major feature is Entire’s distributed Git hosting infrastructure, which allows developers to create, migrate and host private code repositories on the platform.
The company claims that its system can handle code pushes at speeds up to 89 times faster than GitHub and six times faster than Cursor Origin, a competing code-hosting service launched by AI coding company Cursor.
These figures are based on Entire’s own benchmarks and should be understood as company-reported performance claims rather than independently established results across all workloads.
The platform supports private repositories in the United States, Europe, India and Australia. Regional hosting can help development teams manage where their code is stored, an important consideration for organisations with data-residency requirements.
Developers who are not ready to leave GitHub can continue using their existing repositories while mirroring them to Entire. Teams can also move repositories to Entire’s native hosting service.
The company plans to expand its platform with additional capabilities, including public repositories and self-hosted Git infrastructure.
Continuous Memory helps AI agents learn from previous work
Entire’s third major feature, Continuous Memory, is designed to address another limitation of AI coding tools: the loss of useful context between sessions.
When an AI agent works on a software project, it may make decisions based on previous instructions, debugging sessions and changes to the codebase. If that context is unavailable during a later session, the agent may repeat earlier mistakes or spend additional resources rediscovering information.
Continuous Memory stores coding-session information and makes it searchable within the repository. This allows AI agents to retrieve relevant context from earlier work instead of starting each task without access to that history.
The system includes a codebase graph that helps agents understand relationships within a project, alongside a memory component that records previous agent sessions and prompts.
Entire claims that its codebase graph reduced token usage by as much as 71% in its testing. Tokens are units used to process text and other information in AI systems, and reducing token consumption can lower costs and improve efficiency. The reported reduction is a company claim and may vary according to the project and workload.
Partnerships aim to speed up testing and deployment
Generating code is only one stage of software development. The code must also be built, tested and checked before it can be released.
To address this requirement, Entire has partnered with Buildkite and Depot, providers of software build and testing infrastructure.
The integrations allow development teams to connect their existing workflows with Entire’s review system. Build and test jobs can run when code is pushed, with the results appearing in Trails.
The aim is to ensure that the processes following code generation do not become a bottleneck as AI agents produce more changes.
The partnerships also provide a transition path for teams that currently use GitHub Actions. Entire says developers can convert existing workflows for use with the partner platforms, allowing them to retain parts of their established development processes while adopting the new system.
Entire has raised $60 million in seed funding
Entire’s latest launch builds on its earlier funding and product development.
In February 2026, the company announced a $60 million seed funding round led by Felicis, with participation from investors including Madrona, Microsoft’s venture arm M12 and Basis Set Ventures. The company was valued at $300 million at the time of that announcement.
Other backers included technology industry figures such as Yahoo co-founder Jerry Yang and Y Combinator chief executive Garry Tan.
Entire initially introduced Checkpoints, an open-source command-line tool that records AI coding agents’ prompts, decisions and session context alongside code changes. The company subsequently developed its distributed Git network before unveiling the broader platform.
The startup has grown to a global team of approximately 50 employees, according to reports published around the launch. Its team includes people with experience at technology companies such as GitHub, Microsoft and Atlassian.
Can Entire compete with GitHub?
Entire enters a competitive market in which established software platforms and AI companies are rethinking how developers create and maintain code.
GitHub benefits from a large developer community, established enterprise relationships and extensive integrations. Microsoft also has the resources to continue investing in its development tools and AI capabilities.
Entire’s strategy is to target a changing need rather than compete solely on conventional code hosting. Its emphasis on automated reviews, persistent agent memory and faster infrastructure is aimed at teams that use AI agents extensively.
The platform also faces challenges. Automated evaluations cannot guarantee that every change is correct or secure, and organisations may still require detailed human review for critical software. Developers will also need to assess performance claims, security arrangements, pricing and the effort involved in moving existing projects.
For Entire, success will depend on whether its tools can make AI-driven development more reliable and productive while earning the trust of individual developers and large organisations.
Dohmke’s new venture reflects a broader shift in software engineering: as AI agents take on more coding work, the tools used to manage, review and deliver that work may need to change as well. Entire is betting that the next generation of developer platforms will be built around collaboration between humans and AI agents rather than adapting traditional workflows alone.