In the rapidly evolving landscape of artificial intelligence and software development, a new player has emerged that promises to bridge the gap between human intent and machine-generated code. Nat Friedman, the former CEO of GitHub who steered the platform through its acquisition by Microsoft and the rise of AI-assisted coding tools like Copilot, has launched Entire — a groundbreaking developer platform designed specifically for collaboration between humans and AI agents.
With a massive $60 million seed funding round, Entire positions itself as the "world's next developer platform," aiming to redefine how developers work in an AI-native world.
The Genesis of Entire
Friedman, who left GitHub in 2023 after seven years at the helm, has long been at the forefront of AI's integration into software engineering. His latest venture, co-founded with a team of innovators, addresses a critical pain point in AI-driven development: the loss of context and reasoning behind code changes. Traditional tools like Git excel at version-controlling source code, but they fall short when it comes to capturing the interactions with AI models that increasingly produce that code.
Entire's core innovation is an open-source command-line interface (CLI) tool that extends Git's capabilities. This tool allows developers to store not just the code, but the entire history of communications with AI agents—prompts, session transcripts, files touched, token usage, and tool calls—as structured metadata. These "Checkpoints" are tied to commit SHAs, providing a comprehensive audit log of how and why code was generated.
How Entire Works: Checkpoints and Git Integration
At its heart, Entire reimagines Git as a unified database for code, intent, constraints, and reasoning. When an AI agent generates a commit, the CLI automatically captures the session details and pushes them to a dedicated branch called "entire/checkpoints/v1." This append-only structure ensures that the metadata remains organized and doesn't clutter the main commit history or comments.
Why a separate branch instead of embedding data in commit messages? The choice likely stems from practicality and scalability. Commit comments are meant for human-readable summaries and could become unwieldy with large volumes of structured data like full transcripts or token logs.
A dedicated branch keeps the primary codebase clean while allowing easy querying and reference — ideal for AI agents that need to access past contexts without parsing noisy commit logs. This design also supports multi-agent workflows, where different AIs can build on each other's work without repeating errors.
Setup is straightforward: Developers can install the tool via a simple curl script and enable it in any Git repository with the command `entire enable`. Once active, it hooks into pushes, syncing AI sessions seamlessly.
The Entire Platform: Convenience and Collaboration
Beyond the CLI, Entire offers a user-friendly service at entire.io, making it easier to manage and leverage these checkpoints. The platform provides a terminal-based interface for searching commit histories, viewing agent statuses, and diving into session details — all without leaving your existing workflow. It's designed to eliminate context switching, ensuring that developers stay in their preferred tools while gaining AI-specific insights.
Key features include:
- Traceability: Easily track changes made by AI, complete with the reasoning behind them.
- Faster Reviews: Reviewers can see the full intent and constraints, speeding up pull requests.
- Better Handoffs: Teams can pick up where an AI left off without replaying entire sessions.
- Efficiency Gains: By referencing past work, agents avoid redundant mistakes, reducing token waste and improving overall productivity.
- Multi-Session Support: Enables complex, ongoing projects involving multiple AI interactions.
Entire is fully open-source under the MIT license and free to use, with no additional pricing tiers mentioned — making it accessible for individual developers and enterprises alike. Community engagement is encouraged through GitHub Discussions and a Discord server, fostering rapid iteration based on user feedback.
Fueling the Vision: $60 Million Seed Round
To bring this ambitious vision to life, Entire secured a staggering $60 million in seed funding, led by Felicis Ventures and supported by a roster of heavy-hitters including Madrona, Microsoft's M12, Basis Set, 20VC, Cherry Ventures, Picus Capital, Global Founders Capital, and individual investors like Gergely Orosz, Theo Browne, Jerry Yang, Olivier Pomel, and Garry Tan. This investment underscores the industry's belief in Entire's potential to become the go-to platform for AI development.
Friedman and his team argue that as AI agents become primary code producers, traditional tools like issues, pull requests, and even Git itself—optimized for human collaboration — need an overhaul. Entire steps in to provide durable context, enabling agent interoperability and a new software development lifecycle tailored for this era.
Why This Idea Matters: Preserving Human Essence in AI Code
The concept behind Entire is profoundly sensible. Interactions with AI models represent the human-directed essence of a project—the prompts, constraints, and iterations that shape the final output. Storing these alongside the code ensures that this vital context isn't lost in ephemeral chat histories or discarded sessions.
One standout benefit is empowering AI itself: Agents can reference their own project history, learning from previous outputs to avoid repeating mistakes. This creates a feedback loop that enhances reliability and efficiency in AI-driven workflows. In a world where AI is increasingly autonomous, such traceability could mitigate risks like unintended biases or errors propagating through codebases.
Whether Entire becomes the dominant platform remains to be seen, but its foundational idea—treating AI communications as first-class citizens in version control — feels like a natural evolution. It shifts the focus from mere code generation to holistic system-level thinking, addressing bottlenecks in context and architecture that Friedman identifies as key challenges.
Also read:
- Key Updates to the Quasa Connect App for Crypto Freelancing
- Elon Musk's X Launches Crypto and Stock Trading: Turning the Timeline into a Trading Terminal
- BlackRock Enters DeFi: World's Largest Asset Manager Lists $2.2B Tokenized Treasury Fund BUIDL on Uniswap
Get Started Today
Entire is available for use right now. Head to the GitHub repository at https://github.com/entireio/cli to download the CLI, or visit entire.io for more details on the platform. As AI continues to transform software engineering, tools like Entire could very well become indispensable, ensuring that the human spark behind the machine remains preserved and actionable.
