Harrison Chase

Deep agents Deep agents Deep agentsThomas Taylor: learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents Link: https://x.com/mrthomastaylor/status/1989128654127866224

RT LangChain Our very own @RLanceMartin outlined a new playbook for AI engineering on the High Signal Podcast. In this conversation, he touches on: 🔶 Why top products from Claude Code to Manus are constantly re-architecting to keep up with tomorrow's models 🔶 How to use context engineering to manage cost & reliability 🔶 Why evaluation for agents is critical (and why you should start simple) 🔶 The power of an "agent harness" to extend AI capabilities without adding complexity .... And more! Check out the episode in the comments to hear more from @dsgilchrist @hugobowne and @RLanceMartin

RT george salapa 🜁 i really like this mental model of what langgraph bc it clicked for me graph is a looping mechanism with predetermined routes (ex: interruption for human approval) state is a dict that travels through the loop node are basically tools

RT Jake Broekhuizen Great overview from prof. @RLanceMartin on context engineering and how principles like offload, reduce, and isolate show up in modern agent architectures - including our new Deep Agents package + CLI. https://www.youtube.com/watch?v=XFCkrYHHfpQ

RT Thomas Taylor learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents learn about deep agents

RT Christian Bromann ⚡️🚀Mason Daugherty: Hour zero support for GPT-5.1 in your @LangChainAI agents! Link: https://x.com/masondrxy/status/1989053478321439052

RT Mason Daugherty Hour zero support for GPT-5.1 in your @LangChainAI agents!Sam Altman: GPT-5.1 is now available in the API. Pricing is the same as GPT-5. We are also releasing gpt-5.1-codex and gpt-5.1-codex-mini in the API, specialized for long-running coding tasks. Prompt caching now lasts up to 24 hours! Updated evals in our blog post. Link: https://x.com/sama/status/1989048466967032153

RT LangChain ⭐️New LangChain Academy Course: LangSmith Essentials⭐️ Testing applications is essential to the development lifecycle, but LLM systems are non-deterministic – you can’t always predict how they will behave. Add multi-turn interactions and tool-calling agents, and testing agents becomes even more complex than traditional software testing. To address this challenge, LangSmith provides a comprehensive platform for agent engineering that helps teams use live production data for continuous testing and improvement. In our new quickstart course, LangSmith Essentials, you’ll learn how to observe, evaluate, and deploy your agent in less than 30 minutes. Enroll for free ➡️https://bit.ly/4hYBVq1

RT Sydney Runkle 🎬 Day 3 of our LangChain middleware series! ✅ Today: To-Do List Middleware Your @LangChainAI agents can now plan like Claude Code does, breaking down complex tasks and checking them off as they go. Add this in with just 2 lines of code! https://www.youtube.com/watch?v=yTWocbVKQxw

RT Christian Bromann Your AI agent keeps looping, forgetting steps, or losing context? There’s one simple trick to make it smarter: TodoListMiddleware 🧠 It gives your LangChain agent an actual plan — so it can think ahead, track progress, and finish what it starts. 🎥 Watch now → https://youtu.be/dwvhZ1z_Pas #LangChain #AI #Agents #LLM #TypeScript #Nextjs

RT Viv Re this was a ton of fun to work on, sandboxes are quickly becoming a default key part of agent infra. lots of use cases like safe code exec, reproducible work, and massively parallel jobs. hype to see what the community builds around this, reach out with any questions from the code, vid, or blog :)

RT LangChain 💻Sandboxes for DeepAgents We're excited to launch Sandboxes for DeepAgents, a new set of integrations that allow you to safely execute arbitrary DeepAgent code and bash commands in remote sandboxes. Supports @RunloopAI @daytonaio @modal Your DeepAgent runs locally (or wherever you want), but when it needs to execute code, create files, or run commands, those operations happen in the remote sandbox. Blog: http://blog.langchain.com/execute-code-with-sandboxes-for-deepagents/ Docs: https://github.com/langchain-ai/deepagents Video: https://youtu.be/CejntUP3muU

RT LangChain The AMA with @hwchase17 starts in 30 mins! 🎉 Ask about LangChain, LangGraph, or the future of agents, async on the forum (no cameras 👀). Drop your Qs now or join once we start 👇 🔗 https://forum.langchain.com/t/im-harrison-chase-ask-me-anything-on-nov-13-14th/2113 🕗 Nov 13–14 | 8AM–6PM PST

RT Christian Bromann Ever wish your AI agent would ask before acting? 🤔 I built a Human-in-the-Loop middleware for @LangChainAI that pauses execution until you approve the next step. 🎥 Watch the full demo → https://youtu.be/tdOeUVERukA 🧑‍💻 @nextjs demo app → https://github.com/christian-bromann/langchat #LangChain #AI #Agents #Nextjs #TypeScript

RT LangChain Context Engineering in Practice A few common agent design principles have emerged -- offload/reduce/isolate agent context. Here, we cover examples of each across popular agents, including our own open source deepagents harness. 📹: https://youtu.be/XFCkrYHHfpQ

RT LangChain Most agents fail at complex tasks because they lose context or can’t plan ahead. Deep agents are architected differently to handle long-running, multi-step workflows (like Claude Code). Deep agents have four characteristics that differentiate them from typical looping agents. While they still use the basic agent framework of an LLM calling tools in a loop, the capacity for deep work is enabled through: 🧠 A detailed system prompt 📋 A planning tool 🤝 Subagents 🗂️ A file system tool In this post we discuss each of these building blocks and how they enable agents to go deep: https://blog.langchain.com/deep-agents/

RT LangChain 👩Lang-ladies NYC Breakfast☕️ Join us Nov 20 for breakfast + coffee with women in the LangChain community! Connect, collaborate, and talk agents and real-world AI. 📍USQ area ⏰8:15–9:45am ET 🎟️ Space is limited — RSVP here to confirm your spot: https://luma.com/zvou4qkv For women & gender minorities in tech 💪

RT LangChain Excited to see FlowAgent by @TearlineAI live — it's built on LangChain and LangGraph, and helps folks orchestrate complex Web3 tasks.Tearline: @TearlineAI × @LangChainAI DeFi meets AI. Powered by LangChain, our flagship FlowAgent makes multi-agent workflows visual, modular, and maintainable, with built-in persistence and cost-efficient execution. We built a PDCA Agent that plans, checks, and delivers verified DeFi Link: https://x.com/TearlineAI/status/1987931427867111785

RT Hamel Husain This eval talk features some of my favorite people all in one go. It's discusses evals from many perspectives: - How to look at data - Human/Computer interface design - Metrics - Tools - etc @eugeneyan , @sh_reya , @BEBischof , @hwchase17 , etc 🔥 https://www.youtube.com/watch?si=P9EmuJXw0kzLsdIu&v=SnbGD677_u0

RT Warden Calling All Agent Builders 👾 The Warden Agent Hub opens this month: your gateway to 13M+ users and a $1M incentive pool for AI & DeFi agents. ⚡ Publish in minutes via Warden Studio 🌍 Join the first wave of onchain AI agents ✅ Early Onboarder Bonus (Top 10 agents published within the first month earning $10,000 each) 💸 Monetize instantly in USDC

RT Amada Echeverría 👩🗽Join us for an intimate breakfast with fellow women in the LangChain community! 🥐 https://luma.com/zvou4qkv

RT Edwin Estrada Langchain team overcoming all the objections and haters head on Lang* v1 is 🔥🔥🔥 Community thanks you 🙇 @hwchase17 @sydneyrunkle @mrthomastaylor breaking down the pain points solved ftw https://www.youtube.com/watch?v=SX_h7bUBBww

RT LangChain 📊🤖 Stock Research Agent V3 (Made by the LangChain Community) An AI platform transforming financial research, leveraging LangGraph and LangSmith for agent collaboration and real-time monitoring while delivering 73% cost savings. Check it out on GitHub! 🚀 https://github.com/sagar-n/deepagents/tree/v3.0.0/deep-research-agents-v3

RT LangChain 📅🤖 Event Deep Research (Made by the LangChain Community) An AI-powered system that generates historical timelines using LangGraph's multi-agent architecture. Automatically researches and compiles comprehensive JSON timelines of significant life events. Explore this innovative tool! https://github.com/bernatsampera/event-deep-research

RT freeCodeCamp.org If you're building AI apps, you might be struggling with how to structure the logic and connect prompts, memory, tools, and APIs. Well, this is where open-source tools like LangChain and LangGraph can help you out. In this guide, Manish explains what each tool does and when (and how) to use it in your AI projects. https://www.freecodecamp.org/news/how-to-use-langchain-and-langgraph-a-beginners-guide-to-ai-workflows/

RT Daniel There's a lot in useStream. Gotta make good use of it!Harrison Chase: Deploying LangChain agents in NextJS! For folks looking to both build and deploy an agent in the NextJS ecosystem - this is a great resource to get started! Link: https://x.com/hwchase17/status/1986491837889323451

🚀Mini "TypeScript AI" release day! We released a bunch of things in the Lang* ecosystem to make building AI agents in TypeScript easier than ever: 🤖DeepAgents 1.0: https://docs.langchain.com/oss/javascript/deepagents/quickstart 🕸️Tutorial on building agents with NextJS: https://youtu.be/piK5WTXAEAQ 📃Tutorial on building a deep research agent with DeepAgents TypeScript: https://youtu.be/mUNeBCtJKk0

RT Kevin Fischer Kids deserve better screentime So we built Spark: world-class puppeteers + AI performing a character who actually knows your family Real humans ensuring real connection. Think Sesame Street meets the future First 20 Portals shipping this month Interviewing the next 200 families now

RT Sydney Runkle My colleague @bromann is cranking out fantastic educational content re: building agentic applications with @LangChainAI in typescript!Christian Bromann: New video: Build a streaming @LangChainAI agent in @nextjs using useStream + memory 🚀 You’ll learn: - stream AI replies into your UI with useStream - Minimal API route serving SSE - Add conversation memory via thread id + checkpointer 🎥 Watch now: https://www.youtube.com/watch?v=piK5WTXAEAQ Link: https://x.com/bromann/status/1986491398665929209

RT Pawan Built a VS Code extension on top of LangChain’s DeepAgents — shows AI code changes inline with human approvals + command execution. Next: adding green/red highlights diff and accept/reject buttons per line. What’s your biggest pain point with AI coding tools?

RT LangChain SUPER EXCITED to be working with @privy_io on wallets to enable agentic commerce! fun collab between @MaxSegall and @j_schottenstein -- crypto + ai agents in 1 household coming together for a partnership 🤝Privy: 1/ AI agents can now pay with stablecoins. We’ve teamed up with @LangChainAI, the leading agent framework, to seamlessly provision wallets for agents so they can transact onchain. Stablecoins are programmable money. Privy makes them safe for agents to use. Here's how 👇 Link: https://x.com/privy_io/status/1986503492547039502

RT Julia Schottenstein A @privy_io @LangChainAI collab for the ages ♥️Privy: 1/ AI agents can now pay with stablecoins. We’ve teamed up with @LangChainAI, the leading agent framework, to seamlessly provision wallets for agents so they can transact onchain. Stablecoins are programmable money. Privy makes them safe for agents to use. Here's how 👇 Link: https://x.com/privy_io/status/1986503492547039502

Video walkthrough of building a deep research agent - in typescript!LangChain: 🧑‍🔬 Building a Typescript deep research agent In this video, we will walk through how to easily build a Typescript deep research agent This builds upon our new DeepAgents library All it involves is some detailed prompting, some search tools, and some specialized sub agents Link: https://x.com/LangChainAI/status/1986503307578474668

RT LangChain 🧑‍🔬 Building a Typescript deep research agent In this video, we will walk through how to easily build a Typescript deep research agent This builds upon our new DeepAgents library All it involves is some detailed prompting, some search tools, and some specialized sub agents Check out our video here: https://www.youtube.com/watch?v=mUNeBCtJKk0

RT LangChain Excellent walkthrough of how to build a streaming agent in @nextjs using LangChain. Clear explanation of server-sent events, UI streaming, and memory via thread IDs. If you’re evaluating production agent architectures, this is a strong reference implementation. 👇 Watch the full video here: https://www.youtube.com/watch?v=piK5WTXAEAQ 🧑‍💻 Clone the repository and give it a go: https://github.com/christian-bromann/langchain-nextjsChristian Bromann: New video: Build a streaming @LangChainAI agent in @nextjs using useStream + memory 🚀 You’ll learn: - stream AI replies into your UI with useStream - Minimal API route serving SSE - Add conversation memory via thread id + checkpointer 🎥 Watch now: https://www.youtube.com/watch?v=piK5WTXAEAQ Link: https://x.com/bromann/status/1986491398665929209