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

AI 服务暂时不可用，以下为来源正文，待恢复后补全翻译。

Newsletter August 2026: LangChain Newsletter August 26, 2026 4 min Go back to blog Create agents This month, Harrison talks about why owning your intelligence matters more than ever. Also new: Managed Deep Agents and LLM Gateway hit public beta. Keep on reading to see what's new at LangChain. LangSmith Updates 🤖 Managed Deep Agents Public Beta: Deploy your Deep Agent to a managed LangSmith runtime with one command, with durable execution, sandboxes, tracing, and more built in. 🔐 LLM Gateway Public Beta: LLM Gateway sits between your agents and the models they call, providing the controls that teams need first when agents go into production: cost controls, rate limits, model fallbacks, and sensitive data handling. 🏎️ LangSmith Engine Performance Improvements: Engine is now more than 2x better at finding agent issues, with proposed fixes that perform 25% better on standard benchmarks. ⚖️ Tuned Evaluators: Automatically attach quality feedback to production traces, so you can find and fix agent mistakes without managing your own judge. Our first Tuned Evaluator is Perceived Error. ☁️ Bring Your Own Cloud: BYOC is now available on AWS, so teams can run LangSmith in their own VPC while LangChain handles setup, scaling, and upgrades. Open Source 🔧 Deep Agents v0.7: This release simplifies the base harness, resulting in 65% fewer base input tokens at comparable performance. What does it mean to "own your intelligence"? Over the next five years, every company will use AI in one of two ways: to run critical parts of their business, or as part of the product they sell to customers. In both cases, generic intelligence will not be enough. In his latest blog, Harrison talks about why companies need to own their intelligence to have real impact. Read the blog→ Speak the Lang New Series: Managed Deep Agents In this video series, Harrison provides an overview of Managed Deep Agents, covers core concepts, and a quickstart to create and deploy your first Managed Deep Agent. Watch now → New in Foundation: Introduction to Deep Agents We've updated this course with material around the Deep Agents v0.7 release, plus added new subagent modules, support for Typescript, a built-in tutor skill, and more. Enroll now → Upcoming Events 🤖 Silicon Valley · Dallas · Boston · Los Angeles LangSmith Roadshow: Building Agents with Agents ‍Join us for a half-day of technical sessions, live demos, an engineering leader spotlight, and a hands-on lab focused on building and improving agents in production. ‍Find your city → ‍ 🗽🇬🇧 Sep 24 & Oct 13 Interrupt is coming to NYC & London Hear keynotes from AI industry leaders, network with fellow builders, and learn how to ship and improve your agents. Find your city → ‍ 🔃 San Francisco · Sep 2 ‍LangChain Presents The Learning Loop: A summer event on continual learning RSVP → ‍ 🤖 Hamburg, Germany · Sep 3 LangChain Meetup: Building Great Agents with Accenture & Jimdo RSVP → ‍ 🇸🇪 Stockholm · Sep 8 LangChain x Brickanta Meetup: The AI stack behind the Swedish Startup Company of the Year RSVP → ‍ 🇫🇷 Paris · Oct 5 Building Agents Meetup with LangChain & Pigment RSVP → ‍ 🌷 Amsterdam · Oct 6 LangChain Amsterdam Meetup with Harrison Chase RSVP → ‍ Attending COLLIDE in Atlanta? Find LangChain at the conference on Oct 1. What teams are building with LangChain Stripe built Kai, a company-wide productivity agent, on Deep Agents in one week, then scaled it from 296 to over 5,000 users in a month. Today, 83% of Stripe uses Kai weekly across 60,000+ sessions. Read the story → Apollo rebuilt its AI Assistant on Deep Agents to handle the full GTM loop in one natural-language chat. The shift shrank new agent development time by roughly 80-85%. Read the story → ‍ How can you follow along with the Lang Latest? Check out our Blog, Changelog, and YouTube channel for more updates. For any additional questions or feedback, email us at [email protected]. See what your agent is really doing LangSmith, our agent engineering platform, helps developers debug every agent decision, eval changes, and deploy in one click. Try LangSmith Get a demo