LangChain expands agent infrastructure: Engine v2, Managed Deep Agents and SmithDB

LangChain rolled out production agent infrastructure: LangSmith Engine v2 automatically reproduces failures and tests fixes, Managed Deep Agents in public beta handles the runtime, and SmithDB speeds up observability workloads up to 15x. Sandboxes on microVMs, LLM Gateway and Context Hub add security and governance.

DEV Community

Open source

LangChain expands agent infrastructure: Engine v2, Managed Deep Agents and SmithDB

Engine v2 reproduces failures and tests prompt and code fixes before human review
SmithDB delivers up to 15x faster performance on core LangSmith workloads
Managed Deep Agents entered public beta in August 2026
Sandboxes run on hardware-virtualized microVMs; LLM Gateway adds spend limits and PII redaction

#LangChain#LangSmith

WSO2 Releases Agent Manager as Enterprises Look to Control Growing AI Agent Sprawl

Security

LangChain expands agent infrastructure: Engine v2, Managed Deep Agents and SmithDB

WSO2 Releases Agent Manager as Enterprises Look to Control Growing AI Agent Sprawl

Eclypsium: attackers shift focus to infrastructure management systems

laya-compactor cuts RAG context tokens by 70% with no answer-quality loss

TypeSafe AI Releases Jev, a Decision-Only Model Returning Typed Probabilities Instead of Text