Building Deep Agents and Deploying in Production
LangChain's open-source Deep Agents adds production harness primitives to any tool-calling agent with one function swap
“if you're not the model, you're the harness”
LangChain introduced Deep Agents, its highest-level open-source abstraction for production agent deployment, built atop LangChain and LangGraph and featuring file-system context management, planning modules, subagent spawning, and custom middleware hooks. It positions as the open-source alternative to closed harnesses like Claude Code, allowing developers to convert existing ReAct agents with a single function call. The talk frames the agent reliability-vs-agency tradeoff as increasingly resolved by improving models and harnesses together.