Anthropic's Flows lets Claude orchestrate up to 1,000 agents from a single goal
AI Engineer · youtube · 2026-10-17
At AI Engineer NYC, Anthropic's CJ Avilla and Harrison Stall introduced Flows in Claude Managed Agents: hand Claude a goal and it plans and runs a multi-agent topology of up to 1,000 agents via API. Key points:
- Why single agents fail: laziness, self-preference bias, goal drift
- Demo: stock watchlist scanner with per-phase compute tuning
- Templates: Bug Hunter (multi-agent PR review) and Explainer Studio (flows that make videos)
- Safety at scale: budgets, interrupts, auto mode, credential vaults, network lockdown, and separating the brain from the hands
More from coding & agent
- LLM one-shots a Minecraft 3x3 piston door by building its own Linux test infra — Vjeux · 2026-10-18
- Dev builds Mumbai-inspired 3D manga background generator from OpenStreetMap data — CurieuxExplorer · 2026-10-18
- Solo founder builds a money-printing data service on Claude's Pi Durable harness — MicahBerkley · 2026-10-18
- Dev builds Arch-based OS with custom desktop and browser using Claude Code — Snoo-10493 · 2026-10-18
- Pi Durable went through 7 rewrites yet agent session files still lack a standard — ricklamers · 2026-10-18
- Five SaaS agent workflows fail the same way — and it's never model intelligence — usehive · 2026-10-18