Entire Inc., a startup led by former GitHub Chief Executive Thomas Dohmke, launched today with $60 million in funding.

Felicis led the seed round with participation from Microsoft Corp.’s M12 fund, Madrona and other venture capital firms. They were joined by several prominent angel investors. The group included Datadog Inc. CEO Olivier Pomel, Y Combinator CEO Garry Tan, former Yahoo CEO Jerry Yang and others.

Software teams manage the code they write using an open-source tool called Git. The software, which powers GitHub and GitLab, stores every single change to an application’s code base. When developers make a new change, the update has to be approved by a colleague before it’s saved.

Entire argues that the traditional way of using Git is not optimal for projects in which engineers use artificial intelligence agents to produce code. Currently, software teams save AI-generated code changes, but not the prompts that developers used to generate them. Those prompts contain valuable information that can speed up application projects.

Entire is developing a Git-compatible platform that saves developers’ prompts. According to the company, its platform also logs other details such as which third-party tools an AI agent used to generate code. Entire’s platform will make that information available to both developers and AI agents.

The company says that its approach can streamline development in several ways. 

When engineers ask an AI agent to generate code and it makes a mistake, they often enter a follow-up prompt with remediation guidance. Entire can save that guidance and make it available to a software team’s other AI agents. Reusing troubleshooting advice enables AI agents to avoid repeating mistakes.

Developers, meanwhile, can use the prompt logs saved by Entire to understand the reasoning behind AI-generated code changes. The logs also reduce duplicate work. Before developers use an AI agent to tackle a technical difficulty, they can review Entire to determine if a colleague may have solved the issue in a past chatbot session. That workflow reduces unnecessary inference costs.

“Just like when automotive companies replaced the traditional, craft-based production system with the moving assembly line, we must now reimagine the software development lifecycle for a world where machines are the primary producers of code,” Dohmke said.

The first component of Entire’s platform is an open-source tool called Checkpoints that it released on GitHub today. The software logs the AI prompts that developers use to generate code along with associated details such as token use. It can collect data from Claude Code and Google LLC’s Gemini CLI on launch with support for more tools set to roll out in the coming months.

In the longer term, Entire plans to release a “semantic layer” that will make it easier for AI agents to use project data saved by its platform. The company is also working on an interface “where developers can review, approve and deploy hundreds of changes” per day. It will double its 15-person team to accelerate its product development efforts.

Photo: Unsplash