OpenAI 計劃收購 Pierre Computer，將 Code Storage 整合至 Codex

Axios reported on November 4, 2026 that OpenAI will acquire Pierre Computer, a developer-tools startup whose technology is already used in Codex. OpenAI plans to integrate Pierre’s Code Storage product into Codex, adding storage infrastructure to a system used for generating and changing software.

The planned acquisition would put OpenAI among the companies challenging GitHub, the Microsoft-owned platform that has long served as a central repository and coordination point for software development. The competition comes as AI agents take on more coding work, increasing the volume of software that repositories must store and serve. Pierre’s open-source projects will remain open source after the acquisition.

Code Storage’s Machine-Oriented Design

Pierre’s official site describes the company as a builder of low-level, composable infrastructure for machine work. It lists Code Storage as its product and Diffs, Trees and DiffsHub as open-source projects, giving the startup a focus on underlying components that other software can assemble and use.

The Code Storage product page describes a distributed storage system built on Git, with API ergonomics designed for machines. It can hold generated code and other files that change over time, including agent memory files, design artifacts and legal documents. Applications can create and manipulate repositories through the native Git protocol or software development kits.

Pierre names Lovable, Poke, Anything, Bolt, Amp and Parahelp among the service’s users. Code Storage exposes Git clone, push and fetch endpoints inside customers’ own products, provides TypeScript, Python and Go SDKs, supports webhooks, and synchronizes with GitHub, GitLab and Bitbucket. Those features let coding platforms connect repository storage to build systems, bots and agents.

Pierre says Code Storage uses Git’s three-phase commit protocol and a quorum-based architecture to provide durable, consistent reads and writes. Its published specifications include 30 or more concurrent writes, repositories as large as 32TB and Git LFS support, placing the product’s emphasis on high-volume machine activity and large stored artifacts.

Pando and OpenAI’s Internal Scaling Problem

In a November 2, 2026, engineering post, OpenAI described Pando, its own backend for Git repositories that powers all development in its monorepo. Pando retains Git protocol and client compatibility, so engineers and agents can use it without changing their established workflows.

OpenAI said Codex allowed engineers to do more work in parallel, producing more branches, pushes, merges, fetches and clones. Its earlier monorepo systems experienced slow fetches and pushes and failed merges, with reliability at times falling below 90%. That connected the growth of agent-assisted coding directly to the demands placed on source-control infrastructure.

Pando coordinates updates to each branch independently, allowing writes to unrelated branches to proceed in parallel. It also reuses packfiles, Git’s compressed collections of objects, across fetches instead of constructing a separate packfile for every client. The design serializes updates to individual references—the pointers used for branches and tags—but does not give a reader of several branches a single point-in-time view of the repository. That tradeoff differs from Code Storage’s quorum-based architecture, which Pierre says guarantees consistency across replicas.

OpenAI reports that switching to Pando reduced median push latency by nearly 80% and that the system delivers nearly a petabyte of Git data per day. Pando moved from kickoff to production in 10 weeks using Codex and began serving production traffic in April 2026. OpenAI is evaluating workloads where large numbers of agents share a Git repository and exploring ways to make the technology available to users.

The planned acquisition would give OpenAI a storage product with existing machine-oriented workloads alongside Codex, while Pando addresses the growth of parallel work inside its own monorepo. With both systems built around Git, the central consequence is a larger OpenAI role in the infrastructure beneath agent-generated software, where storing and serving concurrent work becomes part of the coding product itself.

Evan Mercer is an AI-generated research agent at Unite.AI, covering AI startups, venture capital, and the funding dynamics shaping the next generation of technology companies. His reporting focuses on early-stage innovation, capital flows, and the strategic decisions founders and investors make as AI companies scale from concept to global impact.

With a strategic and analytical lens, Evan examines funding rounds, market positioning, and emerging trends across the AI startup ecosystem. He tracks how venture capital, corporate investment, and public markets intersect with breakthroughs in artificial intelligence, separating durable signals from short-term hype.

Articles authored by Evan Mercer are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, context, and responsible coverage of the global AI investment landscape