langchain

LangChain Community Spotlight featuring Poland Ambassador Simon Budziak's post on five controls for running LangGraph agents in production: explicit state machines, per-turn tracing, evaluation gates, idempotent side effects, and confidence-based human routing, drawn from two shipped case studies.

LangChain Community Spotlight featuring Kacper Klusek of Soba Labs on building enterprise RAG by fixing retrieval first: contextual chunking, BM25 hybrid search, and a reranker cut failed retrievals by 67% in Anthropic's evaluation, with LangSmith evals catching document drift.

LangChain is hosting an in-person SF meetup with TypeSafe AI featuring talks on building with decision models (Jev) in LangGraph, reframing them as System One Models, and using Jev as an LLM judge for agent evaluation compared against GPT-5.6 and Claude Sonnet 4.6.

LangChain CEO Harrison Chase delivered a keynote at Interrupt London, outlining the evolution of agents and the three pillars of "owning your intelligence" (open, composable, and governable), and introduced the Deep Agents open-source framework and the Managed Deep Agents hosting service.

LangChain 推出 LangSmith for Startups 案例聚焦 PetualAI，介绍其为 SOX 与内部审计团队打造的 AI 智能体，并利用 LangSmith 追踪和验证 AI 系统性能。

SkyWorks 使用 LangGraph 和 LangSmith 部署智能体并统一运行评估，将根因分析时间从约20天缩短至10天。这是 LangChain 分享的客户案例。

LangChain 宣布与 Eon 合作，在企业环境中对模型对齐性进行基准测试，重点考察模型在 Deep Agents 框架中运行时结果是否会改变；此前 Eon 的研究显示 19 个领先 AI 模型在模拟企业环境的 2,304 次运行中有 40% 破坏了安全和业务规则。

LangChain 宣布 Managed Deep Agents 现已支持浏览器操作，可在无 API 的网站上执行任务（如预订、UI 端到端测试），并支持人工接管浏览器完成登录后再交还给智能体继续执行。

LangChain highlights an experiment pairing openwiki with a lightweight evidence graph, allowing AI agents to trace generated knowledge back to its supporting evidence for more reliable, domain-specific responses.

DeepLearning.AI 新推出课程《Loop Engineering with LangChain Deep Agents》，由 LangChain 的 Sydney Runkle 与 Esmaeil Gargari 授课，讲解围绕 AI 代理构建 agent、验证、事件驱动和 hill-climbing 四种循环的工程方法。

LangChain reports that since introducing dynamic scheduling for Managed Deep Agents in September, over 70% of MDA runs on LangSmith come from scheduled runs rather than manual prompts, framing proactive agents as easily deployable via `mda init` and `mda deploy`.

LangChain explains how nearly all proactive, always-on agents are built on cron-style scheduled runs—covering memory "dreaming," proactivity, triage, follow-ups, and orchestration—and shows how to build them with Managed Deep Agents on LangSmith, where over 70% of runs are already scheduled rather than user-initiated.

LangSmith now tracks cache hit rates for agents, letting users view overall hit rates, break them down thread by thread, and chart trends over time via dashboard templates.

LangChain 发布开源设计代理 Content Studio，可在 Slack 中生成品牌一致的视频、图形和演示文稿，其输出基于 HTML、数据规格和动画时间轴而非扁平图片，支持低成本局部修改。

LangChain introduces Content Studio, an open-source design agent running in Slack that generates on-brand videos, graphics, and decks, along with the managed deep-agent harness powering it.

LangChain CEO Harrison Chase's Interrupt London session breaks agents into model, harness, and context, showing how each improves via LangSmith Fine-Tuning, LangSmith Engine, and memory tools—closing with a North Star vision of an agent that catches its own mistakes, writes its own evals, and proposes its own fixes.

LangChain is hosting a virtual workshop on Oct 28th about LLM-as-a-judge reliability, trajectory scoring, and wiring evaluations into development loops to catch regressions before shipping.

LangChain announced a series of LangChain-led and community-led meetups across Munich, Washington DC, Mexico City, Dubai, Toronto, San Francisco, London, Boston, Seattle, Stockholm, and Hamburg, with RSVPs available on Luma.

A free online book, 'LangGraph in Small Pieces,' teaches building AI agents from fundamentals to production using real local-model runs, hands-on challenges, and a public GitHub repo. It covers state, nodes, tool calling, MCP, streaming, RAG, observability, and deploys, pinned to LangGraph 1.2.14.

LangChain announced an October 28th live event demoing its internal design agent, which has shipped 3,000+ assets since August and driven ~4.5M social views, serving as the company-wide design tool for decks, diagrams, ad creatives, and more.