OpenAI built Pando for Codex: 5x lower Git push latency, ~1PB transferred daily
sven_ai · x · 2026-11-06
OpenAI revealed that Codex's ability to run more parallel coding tasks strained its original Git infrastructure, so it built Pando in-house.
- Median code push latency reduced by nearly 5x
- Close to 1PB of Git data transferred daily
- Went from kickoff to production in just 10 weeks
The accompanying diagram shows Codex's architecture: each task runs in an isolated sandbox with a dedicated git worktree, orchestrated by a task manager, with human approval checkpoints for network access, out-of-workspace writes, and risky commands, before results are packaged into a PR.
More from coding & agent
- Vercel CEO seeks testers with budget to validate Junior as a market-leading multiplayer coding harness — zeeg · 2026-11-06
- Nubank cuts AI agent update cycles to under five days with simulated customers — astrange · 2026-11-06
- Scyvera open-sources agent governance: scan repos, score blast radius, auto-draft contracts — Trout_dev · 2026-11-06
- How do you catch an agent action that succeeded but was wrong? — krunal_builds · 2026-11-06
- RAME Code: open-source VS Code coding agent runs local Muse Glimmer 30B with permission-controlled tools — acaimari · 2026-11-06
- Open-source tjcoder/cli: terminal agent platform where the model directs its own UI — tjcoder-labs · 2026-11-06