OpenAI details Pando, its Git storage backend that reuses packfiles for CDN-like clone performance
OpenAIDevs · x · 2026-11-03
OpenAI's developer account explained how its Git storage backend Pando works: the Git client reports which commits it has and which it wants, and Pando returns links to stored packfiles covering the gap, including missing branch dependencies. Packfiles may contain objects a client already has, but reusing them across many readers delivers CDN-like performance. The team also contributed fixes upstream to Git, such as cleanup after an interrupted packfile download.
Related event: OpenAI Explains How Pando Fetch Delivers CDN-Level Git Performance(2 posts)→
More from coding & agent
- OpenAI's new Git backend Pando cuts push latency ~80%, serves ~1PB of Git data daily — OpenAIDevs · 2026-11-03
- OpenAI engineer details Pando, the source-control rewrite built for the agent era — OpenAIDevs · 2026-11-03
- Trajectory launches post-training platform with GS1 and Continual Graders research previews — simonguozirui · 2026-11-03
- SMBs are rebuilding their entire GTM stack with agent-native tools, says new thread — garrytan · 2026-11-03
- Compacting Claude Code at 400K instead of 967K cut this dev's cost per request by 43% — Awkward-Let-4628 · 2026-11-03
- Capy 0.7.0 ships Captains orchestration GA; YC's Garry Tan reports 5X issue-fixing throughput — garrytan · 2026-11-03