event · Story package
LangChain says Managed Deep Agents is now in public beta
The vendor describes a managed LangSmith runtime for persistence, tools, sandboxes, traces, approvals, channels, and identity, while the exact beta access terms remain unclear.
Overview
LangChain says Managed Deep Agents is now in public beta as a code-first runtime for deploying Deep Agents through LangSmith. The company describes managed persistence, memory, skills, sandboxes, traces, channels, and identity while developers retain the agent definition.
A related introduction describes durable threads, checkpoints, context, tools, human approval, and traces, but labels the runtime private beta. The supplied evidence does not reconcile the access terms or independently test the service.
Why it matters
A managed runtime could reduce the infrastructure work required to operate stateful agents with sandboxing, approvals, and observability.
Adoption depends on clear availability, permission boundaries, data handling, reliability, and cost, which were not independently evaluated in the supplied evidence.
Key facts
LangChain says Managed Deep Agents is now in public beta for code-first deployment through LangSmith.
The vendor describes managed persistence, memory, skills, sandboxes, traces, channels, and identity while developers retain the agent definition.
A related introduction labels the runtime private beta, leaving the precise access terms unclear in the bounded evidence.
Latest update
No public update is available.
Full timeline
No public timeline entries are available.
Sources
Open questions
- Who can access the runtime, and does the public-beta announcement fully supersede the private-beta description?
- What isolation, permission, approval, and data-retention controls apply in practice?
- How does the managed runtime perform on reliability, latency, cost, and representative agent evaluations?
Related stories
Roundup appearances
No roundup appearances are available.