LangChain adds async subagents to Deep Agents so the main agent can keep talking while work runs in the background
On November 5, LangChain wrote that Deep Agents now support async subagents. An orchestrator stays with the user and keeps conversation context organized, while async subagents run long tasks in the background and report back when they finish. In a standard ReAct loop, the post says, each long tool call blocks the loop and other work queues behind it; splitting the orchestrator from background workers lets the amount of work grow with the number of workers instead of the length of one loop. The post also notes that Claude Code runs subagents in the background by default, OpenAI's dots hand work to background workers, and DeepSeek Harness routes people through a router while workers run tasks in parallel.
Why it matters
Builders of long-running agents get a common pattern—stay responsive to the user while background workers run—in an open-source framework they already use.
Sources1
Written by dalao.ai: the title, summary and why it matters are our own words. The original reports are the record; their publishers keep authorship.