Tag
LangChain recaps its weekly announcements: a new LangSmith Agent Platform, controls for protecting sensitive data in traces, LangSmith for Coding Agents, and the LangSmith Tool Gateway.
LangChain 推出 LangSmith Tool Gateway 公测版,作为托管工具层,为智能体的每次工具调用处理凭证、身份与访问控制,并通过访问、控制、可见性三层治理防止智能体执行未授权或非预期操作。
LangChain is hosting a live virtual workshop on Agent Runtime on November 19th, covering how to build, deploy, and trace agents using LangSmith Managed Deep Agents.
LangSmith introduces a coding agent ROI dashboard that unifies traces across harnesses and model providers, plus plugins for Claude Code, Codex, Cursor, GitHub Copilot, and Deep Agents, with pull requests linked to threads to measure cost per merged PR.
AI agent startup LangChain raised a $300 million Series C led by CapitalG at a $3 billion valuation, more than double its $1.25 billion valuation from a year ago, as it pivots toward its LangSmith agent management platform with 7,000 customers including Nvidia, Toyota, and Salesforce.
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 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 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 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 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 shares a case study on how Numan built Nu, an eval-first AI health coach on LangGraph and LangSmith, navigating medical device regulation by moving beyond static single-turn and scripted multi-turn evals to conversation-aware evaluation frameworks.
LangChain 发布关于 Agent 评估(Agent Evals)的实用指南,建议从真实 trace 中选取 10–20 个失败案例构建小型评估集,结合代码检查与 LLM Judge 逐步完善评估流程。
LangSmith for Startups spotlight features Opine, a customer intelligence platform for revenue teams that traces every production LLM run and gates prompt/model changes behind LangSmith evals, combining real-time data warehousing and semantic search for presales deal context.
Claude Platform traces can now be exported to LangSmith via OpenTelemetry, letting users inspect every step of Claude Managed Agents sessions and Messages API requests. Available in public beta with multiple export paths including server-side destinations, Claude SDKs, and the LangSmith SDK wrapper.
Orange Business 分享其企业 AI 平台 Live Intelligence 如何基于 LangChain 与 LangSmith 构建,涵盖 RAG 质量评估、Agent 行为测试与生产环境可观测性。该平台已在 12 万员工中部署并服务 150+ 客户,选择 LangChain 主要出于模型中立性与欧洲数据主权考虑。
LangChain announces AI Search in LangSmith, allowing users to search through traces conversationally to find answers like user frustration or agent failures instead of manually reviewing hours of data.