LangChain Blog Updates - May 2026
AI · · 1 min read
LangChain Blog Updates (May 2026)
From Token Streams to Agent Streams (May 21)
C. Bromann, N. Hollon. LangChain introduces a new paradigm shifting from token-level streaming to agent-level streaming, enabling real-time agent state observation during execution.
Introducing LangChain Labs (May 14)
Harrison Chase announces LangChain Labs - a new R&D division for experimental agent technologies.
Introducing LangSmith Engine (May 13)
LangSmith Engine - an agent for improving agents. An automated agent evaluation and optimization system that runs as a service within LangSmith.
We built SmithDB, the data layer for agent observability (May 13)
SmithDB is a purpose-built data layer for storing, querying, and replaying agent traces and observability data at scale.
How Auth Proxy secures network access for LangSmith agent sandboxes (May 21)
Enterprise security for agent sandbox environments with network access control via Auth Proxy.
Give Your Agents an Interpreter (May 20)
Hunter Lovell on how giving agents an interpreter improves code execution reliability.
Everything we shipped at Interrupt (May 14)
Major product announcements from the Interrupt conference including:
- LangSmith Sandboxes (GA)
- LangSmith Context Hub
- LangSmith LLM Gateway - runtime governance built into the agent lifecycle
- Managed Deep Agents
- Deep Agents v0.6
- Delta Channels - evolving the runtime for long-running agents
How We Built LangSmith Engine, Our Agent for Improving Agents (May 19)
Deep dive into the architecture of LangSmith Engine - uses agents to evaluate and improve other agents.
The Agent Development Lifecycle (ADLC) (May 9)
Harrison Chase defines the Agent Development Lifecycle - a structured framework for building, testing, deploying, and monitoring agents.