Fchollet on multi-agent coordination: role-based AI teammates running 24/7 in the cloud
fchollet · x · 2026-11-21
François Chollet says the big question of the moment is smoothly coordinating the work of many agents. One approach gaining traction mirrors human organizations: create agents with specific roles — researchers, PMs, testers — each maintaining its own role-scoped context.
He highlights MyClaw 3.0 as an example: cloud-resident AI teammates running 24/7 with built-in tools, shared memory, and support for multiple harnesses (Claude Code, Codex, OpenClaw, etc.), 5,000+ live data sources, and instructions issued from Slack, Teams, Telegram or WhatsApp.
More from coding & agent
- mitsuhiko nudges GitHub to add an AI agent to 4-year-old merge=union feature request — mitsuhiko · 2026-11-21
- One dev orchestrates Fable, Opus and Astra to build a visually striking survival game — lyraxana · 2026-11-21
- JooVoice lets Grok, Muse and other agents make calls in 80+ languages to 67 countries — kimmonismus · 2026-11-21
- CapCut plugin connects ChatGPT to auto-cut gameplay highlights, then human picks the payoff — iamfakhrealam · 2026-11-21
- Dev wires a Runway-generated avatar into Nous Research's Hermes Agent, bashing boring agent UIs — alexcovo_eth · 2026-11-21
- Lean compiles, but the natural-language proof is still wrong — jasondeanlee · 2026-11-21