AI BriefingJune 2026: LangChain Newsletter — Fleet On-Call Copilot, Deep Agents Rubrics, and Other Updates
Key point
LangChain unveiled new features for LangSmith and open-source tools for Deep Agents.
Details
LangChain introduced a range of features to boost agent operational efficiency through a major update to LangSmith. The key updates are as follows.
- Fleet On-Call Copilot: An agent template that analyzes code, traces, and runbooks to triage alarms and draft updates.
- Computer Use in Fleet: Agents can use isolated virtual computers to execute code, manipulate files, and make authenticated API calls.
- Voice Traces UI: Debug voice traces via an audio player and inspect spans at specific points in time.
- Other features: Real-time experiment status tracking and Slack notification integration for LangSmith Engine have been added.
In the open-source area, Deep Agents introduced RubricMiddleware, which helps agents evaluate their own outputs and iterate until they meet the criteria, along with Programmatic Subagents, which run subagents in a distributed manner from interpreter code.
Also newly released are an on-demand course covering the agent development lifecycle (ADLC), a tutorial on adding voice capabilities to LangGraph agents, and an introductory course on LangSmith Deployment.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
LangGraph Agent Server Vulnerability Allows Approval Bypass via State Mutation