OpenAI Codex Lead on Why They Chose Rust and Went Open Source
dotey · x · 2026-09-11
In a Pragmatic Engineer podcast interview, OpenAI Codex lead Thibault (Tibo) explained key decisions:
- Background: Belgian, applied math background, ex-founder and ex-Google/DeepMind. At DeepMind he worked on an internal chatbot predating ChatGPT by a year, but it never shipped; learning ChatGPT had only 20 maintainers drew him to OpenAI, where he joined the o1 preview sprint.
- Why Rust: counterintuitive at the time — models barely supported Rust and peers used TS/Python — but first-principles reasoning said agent cores need robustness, safety and efficiency; compile-time verification fits, and a different language enforced a clean boundary between product UI and agent core.
- Open and model-agnostic: Codex CLI/SDK are open source and support non-OpenAI models, unique among major labs. Rationale: without open source, others could fork with ten lines of changes; benefits include community contributions and winning on experience rather than lock-in. The cost: competitors learn from it before release.
Related event: Codex Lead Explains Why the CLI Is Built in Rust and Open Source(2 posts)→
More from coding & agent
- Dev argues MCP is unnecessary: hand agents an API key and they'll get the job done — sull · 2026-09-11
- Dev to add AI-friendly collision checking CLI to OpenSCAD — _Stocko_ · 2026-09-11
- Cadenya: a year-long indie build of an MCP-powered agent runtime — littlebobbyt · 2026-09-11
- xAI team to livestream building a company from scratch with Grok Bot over three days — soleio · 2026-09-11
- ClawOS: an experimental open-source Arch-based OS with OpenClaw as its primary agent interface — heyneighbor · 2026-09-11
- Grok Bot adds Salesforce, HubSpot, Gong connectors plus installable sales-bot templates — XFreeze · 2026-09-11