LangChain's async subagents: orchestrator stays responsive while background workers run long tasks
hwchase17 · x · 2026-11-08
LangChain's new post "Agents Are Going Async" proposes splitting an agent into an orchestrator that stays available to the user and background subagents that run long tasks and report back on completion.
Key points:
- In a standard ReAct loop, each long tool call blocks everything; async subagents deliver both responsiveness and productivity
- The ecosystem is converging on this pattern: Claude Code runs subagents in background by default, OpenAI's dots hand work to background workers, DeepSeek Harness built its super agent around a router plus parallel workers
- LangChain is bringing the pattern to Deep Agents
More from coding & agent
- Review maps 8 common multi-step agent failure modes to structural architectural fixes — blaizedsouza · 2026-11-08
- Vercel CTO's commit history goes vertical as Opus 4.5 tops his 2016 open-source binge — cramforce · 2026-11-08
- Force sensor + agentic harness one-shots blind precision insertion at industrial reliability — ihorbeaver · 2026-11-08
- Startup cuts observability costs 38x by having Claude Opus migrate off Datadog in one day — LiamFedus · 2026-11-08
- Dutch-Language Bug: A 15k-Char Prompt Made Devs Blame the Model for Their Own Mistake — Initial_Orange2985 · 2026-11-08
- zml team hits "RSI": DeepSeek 4.1 Flash token use exceeds Codex+Claude combined — RemiCadene · 2026-11-08