Summary
In early August 2026, LangChain launched Managed Deep Agents, a service that bundles its Deep Agents harness with managed infrastructure to run production-grade, long-running agents inside LangSmith. It handles the operational hard parts of durable agents: durable execution and checkpointing, streaming, sandboxed tool access, human-in-the-loop workflows, evaluation, persistent memory, and observability.
What changed
LangChain released Managed Deep Agents, a hosted runtime that operationalizes its Deep Agents framework with checkpointing, sandboxes, HITL, memory, and observability built in.
Why it matters
Teams can prototype deep agents quickly but stall on the production plumbing (durability, sandboxing, human approvals, tracing). A managed runtime that ships those defaults shortens the path from demo to production and deepens LangChain's move from open-source framework to hosted platform.
Evidence excerpt
Managed Deep Agents is a hosted runtime for running production-grade deep agents within LangSmith, handling durable execution, checkpointing, streaming, sandboxed tool access, human-in-the-loop workflows, and observability.