deniz.in

Markets

Weather

Loading weather

· via dev.to (home feed)

OpenAI to Acquire Pierre Computer and Bring Git-Based Code Storage Into Codex

OpenAI has agreed to buy Pierre Computer and fold its Git-based Code Storage service into Codex, extending the coding tool toward storage and review infrastructure built for AI agents.

OpenAI to Acquire Pierre Computer and Bring Git-Based Code Storage Into Codex

The acquisition

OpenAI said on Nov. 4 that it has agreed to acquire Pierre Computer, a startup whose Git-based Code Storage service is set to be folded into Codex. The deal remains subject to customary closing conditions, including regulatory approvals, and the two companies will operate independently until it closes. Axios reported that the financial terms were not disclosed.

According to The Daily Flare, the acquisition is meant to push Codex beyond code generation and into the surrounding systems developers use to store and review software, with a focus on adapting Git workflows for teams running multiple coding agents in parallel. OpenAI framed the goal as making Codex a place where developers and agents can work together on large software projects, and connecting the code a business ships with the decisions made while building it.

Pierre's tools are already in Codex

Some of Pierre's technology is already part of the Codex app. Its Trees library helps developers navigate project files, while Diffs shows code changes for review and includes an Edit mode that lets a reviewer fix a line directly where a problem appears. Pierre's recently released Highlights is a syntax highlighter built to update quickly, whether a developer is typing or an agent is streaming code.

Code Storage at the center

The main prize is Code Storage, which Pierre describes as a distributed, Git-based system with an API designed for machine use. AI Weekly reported that the service supports repositories of up to 32 terabytes and more than 30 concurrent writes, and includes Git Large File Storage, SDKs for TypeScript, Python and Go, webhooks, and synchronization with GitHub, GitLab and Bitbucket. The same outlet reported that AI-focused companies including Lovable, Bolt, Amp, Parahelp, Poke and Anything already use the service.

OpenAI said customers can continue using Code Storage after the acquisition closes while its features are brought into Codex. It also committed to keeping Pierre's open-source projects, which AI Weekly listed as Trees, Diffs and DiffsHub, open source, with continued contributions from OpenAI.

The Pando connection

The deal follows OpenAI's unveiling of Pando, an in-house source-control backend for its engineers and coding agents. OpenAI said Pando cut median code-push latency by nearly 80 percent. AI Weekly reported that the system handles nearly a petabyte of Git data each day and reached production in about 10 weeks, with Codex agents assisting the migration work. Axios reported that OpenAI is exploring ways to make Pando available to customers.

OpenAI has not said whether Pando and Code Storage will be combined. Its stated ambition is broader: advancing source control across the industry, from the creation of an agent's first branch through deployment.

A step toward GitHub's territory

Axios noted the competitive dimension of the deal. Microsoft owns nearly a quarter of OpenAI and also owns GitHub, and companies are racing to serve developers as AI agents take on more coding work. Critics have meanwhile raised concerns about GitHub's reliability and about its design being oriented toward human users.

Why it matters

The acquisition signals that AI coding tools are outgrowing the patch-by-chat model. Agents working in parallel on large codebases need branching, storage and review systems built for machine traffic rather than human reviewers, and that is what Pierre built. By buying rather than partnering, OpenAI is assembling its own developer platform layer, shifting from model provider toward infrastructure. With an internal backend in Pando and a customer-facing service in Code Storage, how OpenAI chooses to unify the two will say a great deal about how Codex scales for large software projects, and about the pressure facing incumbent code hosts.

  • #openai
  • #acquisition
  • #developer-tools
  • #git
  • #ai-agents

Related posts