August 2026: LangChain Newsletter — Managed Deep Agents, LLM Gateway, and More

LangChain's August 2026 newsletter announced that Managed Deep Agents and LLM Gateway have entered public beta. Managed Deep Agents let teams deploy a Deep Agent to a managed LangSmith runtime with one command, with durable execution, sandboxes and tracing built in. LLM Gateway sits between agents and the models they call, providing cost controls, rate limits, model fallbacks and sensitive data handling.

LangSmith Engine is now more than 2x better at finding agent issues, with proposed fixes that perform 25% better on standard benchmarks, according to the newsletter. Tuned Evaluators automatically attach quality feedback to production traces, so teams can find and fix agent mistakes without managing their own judge; the first Tuned Evaluator is Perceived Error. Bring Your Own Cloud is now available on AWS, allowing teams to run LangSmith in their own VPC while LangChain handles setup, scaling and upgrades.

In open source, Deep Agents v0.7 simplifies the base harness, resulting in 65% fewer base input tokens at comparable performance.

The newsletter also noted a blog by Harrison on why owning your intelligence matters, a video series on Managed Deep Agents, and an updated Foundation course covering Deep Agents v0.7 with new subagent modules, TypeScript support and a built-in tutor skill. Upcoming events listed include LangSmith roadshows in Silicon Valley, Dallas, Boston and Los Angeles, Interrupt in New York and London on September 24 and October 13, and meetups in San Francisco, Hamburg, Stockholm, Paris and Amsterdam.

Customer stories cited Stripe's Kai, built on Deep Agents in one week and scaled from 296 to over 5,000 users in a month, with 83% of Stripe using it weekly across 60,000+ sessions, and Apollo rebuilding its AI Assistant on Deep Agents, shrinking new agent development time by roughly 80-85%.