OpenAI Announces Plan to Acquire Pierre Computer, Whose Trees and Diffs Libraries Codex Uses, Two Days After Detailing Its Pando Git Backend

Overview

OpenAI announced on November 4, 2026 that it will acquire Pierre Computer Company, Inc., SiliconANGLE reported, adding that the terms of the deal were not disclosed. OpenAI’s announcement describes Pierre as the company behind Code Storage and a family of open-source developer tools, and says that anyone who has browsed files or reviewed a change in the Codex app has already used Pierre’s Trees and Diffs libraries. OpenAI says the closing is “subject to customary closing conditions, including receipt of regulatory approvals”, and that until then the two companies “will remain separate and independent companies”.

The announcement came two days after OpenAI published an engineering post, dated November 2, on Pando, an in-house source-control backend. OpenAI’s acquisition post refers to it as “our in-house source control backend for engineers and agents”.

What OpenAI says it is acquiring

According to OpenAI, Trees lets developers navigate a project’s files and Diffs shows what changed. The post also names Highlights, which it calls “an extremely fast syntax highlighter that updates as a developer types or an agent streams code”, and describes Code Storage as “a service that lets apps store code and track its history using Git”. SiliconANGLE describes Code Storage as a cloud service that AI agents can use to store the code they generate. DevOps.com reported, citing Pierre, that the platform supports repositories as large as 32TB and at least 30 concurrent writes.

OpenAI states several commitments in the same post:

• Pierre’s open-source projects “will remain open-source”, and OpenAI says it will keep contributing to them.
• Customers can keep using Code Storage after the acquisition closes “while we work to bring its features and functionality into Codex”.
• Once the deal closes, OpenAI’s stated ambition is “to advance the future of source control across the industry”.

Pando, described two days earlier

The Pando post, by Aaron Friel, Taylor Blau and Ted Nyman, gives OpenAI’s account of why it built a new backend. It says that starting around the end of last year the systems supporting its monorepo were struggling, with slow fetches and pushes and failed merges, and “reliability at times falling below 90%”. It says Pando now powers all development in the monorepo, that it “cut median push latency by nearly 5x”, and that it delivers “nearly a petabyte of Git data per day”. Later the post says the switch “reduced the latency of the median push by nearly 80%”. By the Herald’s own arithmetic, a fivefold cut and an 80% cut describe the same reduction. All of these are OpenAI’s figures; The New Stack relays the 5x and petabyte figures with the qualifier “the company says”.

On design, OpenAI says Pando coordinates updates to each branch independently and reuses packfiles across fetches instead of generating one for each client. The trade-off it names is that a reader looking at several branches can see them at different points in time. The post says Pando keeps compatibility with the Git protocol and client, and that OpenAI built it with Codex in 10 weeks: the core was designed by one engineer and implemented by teams of agents, and Pando began serving production traffic in April 2026. It was written in Rust.

On availability, OpenAI says it is “currently exploring ways to share the technology that makes Pando possible with our users”. The New Stack reported that OpenAI is not open-sourcing the backend. The Pando post also says several of its Git improvements are already in upstream Git, and many others are in its public repository.

How outlets frame the deal

OpenAI’s two posts, as the Herald read them, do not mention GitHub. The framing comes from the press. SiliconANGLE wrote that observers pegged the deal as a shot at Microsoft’s GitHub for the age of AI agents, and its article does not name them. DevOps.com wrote that the deal strengthens Codex’s position against GitHub as AI coding tools expand into code storage, review and other development workflows.

The New Stack placed Pando among Git-based efforts such as Entire and Cursor’s Origin, and noted that other approaches start from a new client and architecture. On November 8, InfoQ reported on Entire, a Git forge designed for collaboration between human developers and AI coding agents whose CEO is former GitHub CEO Thomas Dohmke, and called it “only one of several new players in the source code management arena”. An earlier Herald analysis, published November 1, compared source-control infrastructure that Cursor, GitHub and other vendors described for agent-driven change volume; it predates both OpenAI announcements.

What is not known

• Terms and timing. SiliconANGLE and DevOps.com both say financial terms were not disclosed. None of the cited sources gives a closing date.
• What moves into Codex, and when. OpenAI says it is working to bring Code Storage’s features into Codex and, per DevOps.com, plans to incorporate Code Storage after the acquisition closes. None of the cited sources gives a timeline, or says whether Pando and Code Storage will be combined.
• Independent measurement. The Pando figures are OpenAI’s own, and none of the cited sources reports a measurement of Pando by anyone outside OpenAI.
• Whether Pando will be offered to customers. OpenAI says it is exploring ways to share the technology; none of the cited sources says what form that would take, beyond The New Stack’s report that the backend is not being open-sourced.