Harrison Chase
@hwchase17
@LangChainAI, previously @robusthq @kensho MLOps ∪ Generative AI ∪ sports analytics
OpenAI recently released a guide on building agents which contains some misguided takes There's a lot of FUD, confusion, hype, and noise around agents I wrote a blog on how to think about agent frameworks. Includes: Background Info - What is an agent? - What is hard about…
super cool workflow for exploring traces
Sharing an overview of the trace reviewer we developed at @FieldDataInc , thank you to the @LangChainAI team for the feedback! Some key points: * soft clustering of traces (promptable) * reasoning model running in bg * hybrid UX for discovery Code and more details in 🧵:…
When Google gutted our pre-training team at Character most people thought we would die—but our numbers don’t lie. 35bn user active seconds per day, 9M Characters created every month, 400% YoY subscriber growth.
I was lucky enough to give a talk at a @LangChainAI last night, arguing that the term "context engineering" is not only here to stay, but signals the emergence of a new field, culture, and community. Here's my write up, if you couldn't make it: dbreunig.com/2025/07/24/why…
Currently a packed room of folks learning about @LangChainAI checkpointers, stateful agents are incoming. Feel free to join, we're live for the next 45 minutes : meet.google.com/jjq-ewsv-szu @hwchase17
Been getting emails about using LangGraph AI agents with local LLMs for privacy. Thinking of adding a chapter to the book on this. Until then, here’s a video on installing Ollama, pulling Llama 3.2 locally, and integrating it with LangChain.js: youtu.be/pKK07kObVnw?si…
📈Open Stock Portfolio Canvas Build a canvas-style agentic application with LangGraph (@LangChainAI) , AG-UI and Gemini 🦜🪁 Featuring: - Generative UI - Canvas UX - LangGraph Agent connected to Yahoo Finance Repo: github.com/copilotkit-sup… Tutorial:…
⚡️Come learn with us at LangChain Academy Live in San Francisco! ⚡️ Experience hands-on learning with LangChain Academy Instructor Lance Martin alongside fellow developers on August 19th, starting at 4pm. We'll dive into the Ambient Agents course, collaborate on building an…
Context Engineering @dbreunig and I did a meetup on context engineering last night. Wanted to share slides (below) + a recap of some themes / discussion points. 1/ Context grows w/ agents. @ManusAI_HQ mentions typical task requires ~50 tool calls. manus.im/blog/Context-E… 2/…
one of the many cool parts about working at @LangChainAI -- we're constantly gathering the community to talk about cool topics. This week was on context engineering!
Full house at the @LangChainAI context engineering meetup Awesome to pickup nuggets of Alpha at these events from those building at the forefront✍️ 🙏@RLanceMartin
Full house at the @LangChainAI context engineering meetup Awesome to pickup nuggets of Alpha at these events from those building at the forefront✍️ 🙏@RLanceMartin
This video by @LangChainAI explains the fundamentals of context engineering in just 14 minutes. Context Engineering is effectively the #1 job of engineers building AI agents. do not sleep on it.
@LangChainAI made it radically easier and more reliable to build multi-step, stateful agents. After training 800+ devs through our @BeyondPrompts workshops, one thing stood out: Checkpointers unlock everything from managing multiple threads with users, to debugging, to…
To those still annoyed by the emergence of a new term – "context engineering" – I recommend this piece I wrote years ago on how buzzwords form, why some succeed (and others fail), and why they matter.
Ambient agents are probably the thing I’ve heard people from LangChain talk about the most! They are agents that listen to event streams and act on them accordingly. Core concepts are the trigger, concurrency and low latency
@dbreunig and I are doing a meetup tmrw on context engineering. Will review own work + common trends / insights we're seeing (e.g., from @ManusAI_HQ, @walden_yan, @kellyhongsn, @barry_zyj + co Anthropic, @dexhorthy). Invite here: lu.ma/comdbqgr x.com/dbreunig/statu…
As your context bloats, you hit different failure modes. These failures hit agents hardest because they operate in exactly the scenarios where contexts balloon: gathering information, making sequential tool calls, engaging in multi-turn reasoning, & accumulating histories.
Some cool new tools from aws!
🛠️ Bedrock AgentCore tools in LangChain It's now easy to use Bedrock AgentCore built-in tools with LangGraph agents. Thank you to our AWS collaborators (github: 3coins and michaelnchin)!
🛠️ Bedrock AgentCore tools in LangChain It's now easy to use Bedrock AgentCore built-in tools with LangGraph agents. Thank you to our AWS collaborators (github: 3coins and michaelnchin)!
How do you build an AI memory system that is used by 10,000+ developers? William will tell you He built @LangChainAI's LangMem He covers: * The flaw of generic memory * How you build a flexible toolkit for devs * And why updates are harder than extraction