LangChain

🔌 New in LangChain MCP Adapters (feat 1/4): multimodal tool support 🌆 MCP tools that return images, text, or other resources now seamlessly convert to LangChain's standard content blocks. 🛠️ This is especially relevant when you expose multimodal agents as MCP tools! Docs: https://lnkd.in/efN2QTFK

💬 Join Sydney Runkle, Chester Curme, and Hunter Lovell on December 15th (December 16th for some APAC time zones) for a collaborative Community Jam Session! We’ll be gathering your thoughts on langchain 1.0 and 1.1.   🦜 🔗 We’ll cover: - langchain 1.0: Now that it’s been out for a while and you’ve had time to explore, we’d love to hear your real-world feedback. - langchain 1.1: We’ll walk through the new features introduced in 1.1 and gather your thoughts if you’ve tried them. - What's next: We'll share more on what we have coming, and recently published updates to langchain-mcp-adapters. Your feedback shapes the future of langchain — see you there! 👉 RSVP: https://luma.com/085nyxmj

🔊 The STT → Agent → TTS “sandwich” is a standard voice agent pattern. It’s easy to get started, tough to build reliable systems. 😵💫 Learn how to debug voice agents: We created a voice agent with Pipecat and sent traces to LangSmith to show exactly how to get visibility into and debug complex pipelines. Stop guessing, start tracing! 👇 Video: https://lnkd.in/g-QJhiEH Docs: https://lnkd.in/grSiQiDn 

Today, we’re shipping new ways to observe, analyze, and debug agents with LangSmith: • Polly: an AI assistant for AI engineering that helps you understand traces, threads, and improve prompts • LangSmith Fetch: a CLI for pulling trace & thread data straight into your terminal or coding agent Agents are running longer and getting more complex, which demands new debugging workflows beyond simple LLM apps. We wrote a blog on the trends behind this shift— and why tools like Polly and LangSmith Fetch are needed. Shipping reliable agents requires full visibility into agent behavior, with tooling that helps you reason over that data in the UI, the terminal, or alongside coding agents. 📔 Learn more about Polly: https://lnkd.in/g6hKM-iM 📔 Learn more about LangSmith Fetch: https://lnkd.in/gG_x7AyP 📔 How to observe deep agents: https://lnkd.in/gwViTXeW 📽️ Polly video tutorial: https://lnkd.in/gEGgbXsU 📽️ LangSmith Fetch video tutorial: https://lnkd.in/gs4JiwCR

Deep Agents represent a shift in how AI systems operate – unlike simple chatbots or basic RAG applications, these agents run for extended periods, execute multiple sub-tasks, and make complex decisions autonomously. In this webinar, we'll dive into practical approaches for gaining visibility into Deep Agent behavior and measuring their effectiveness using LangSmith. RSVP here: https://luma.com/voz2b5h3

✨ Agent engineering: A new discipline ✨ Traditional software assumes known inputs and predictable behavior. Agents give you neither. That’s why teams shipping reliable agents are adopting a new discipline: agent engineering. Agent engineering is driven by a few core ideas: 🔹 Combine engineering, product, and data science skills to shape agent behavior 🔹 Real usage in production shows what actually breaks and where assumptions fall apart 🔹 Agent quality comes from continuous iteration: i.e. ship, observe, refine prompts and tools, repeat Read to learn more about why agent engineering is emerging now — and what it looks like in practice: https://lnkd.in/gmzs_8hg

Voice agents are hard to build. We see two architectures dominate: 1. STT → LLM → TTS (the "sandwich") 2. Speech-to-speech (realtime) Both have trade-offs. The sandwich is model-agnostic, and you can extend existing text agents without rewiring it. But stitching together three separate systems means managing streams, handling interruptions, and fighting latency at every hop. We built a voice agent with LangChain, Cartesia, and AssemblyAI to show how you can build a robust multimodal agent. 📽️ Watch: https://lnkd.in/g3v8qXuC 📖 Docs: https://lnkd.in/ggpb2PdA

🔌 LangChain MCP Adapters 0.2.0 is out! This new release features: 🖼️ Multimodal tool support using LangChain’s standard content blocks ❓Elicitation support via callbacks 🏗️ Structured content for tools, stored as an artifact on tool results 🛠️ Tool name prefixes, preventing name conflicts for multi-server clients 📕 We’ve also rolled out completely new docs: https://lnkd.in/evwGbs-F 📓 Release notes: https://lnkd.in/e_YBiaBb Each day this week I'll do a deep dive on one of these new features, so stay tuned!

Join Harrison Chase and Nick Huang in a deep dive into practical approaches for gaining visibility into Deep Agent behavior and measuring their effectiveness using LangSmith. RSVP: https://luma.com/voz2b5h3

Join Jason Ginsberg (Head of Engineering, Product at Cursor) and Harrison Chase (Co-founder and CEO, LangChain) as they discuss coding agent UX, building with Cursor, and trends heading into 2026. This will be an in-person event in San Francisco! RSVP: https://luma.com/7f7swr1j