OpenAI builds Pando, a Git backend for its monorepo that cuts push latency 80%

AIOpenAI Developers describes Pando, which answers a Git client's request with links to stored packfiles covering the commits it lacks, including missing branch dependencies. Because packfiles are reused across many readers, Pando delivers CDN-like performance even though some returned objects may already be on the client.

AIOpenAI Developers says Pando can update different branches in parallel, while pushes to the same branch are still put in order. On a fetch, each branch tip returns its history, but two tips may reflect different points in time.

AIOpenAI built Pando, a new Git backend for its monorepo that handles more pushes and fetches without changing how engineers and agents use Git. Pando cuts push latency by about 80% and serves nearly a petabyte of Git data a day. OpenAI says the system lets it grow Git capacity alongside agent usage.