hwchase17's profile picture. @LangChainAI, previously @robusthq @kensho

MLOps ∪ Generative AI ∪ sports analytics

Harrison Chase

@hwchase17

@LangChainAI, previously @robusthq @kensho MLOps ∪ Generative AI ∪ sports analytics

Pinned

🧠🤖Deep Agents Simple tool calling loops fail on longer time horizon or more complex tasks Deep Agents like Deep Research, Claude Code & Manus succeed by using a number of tools and tricks We created a new Python package which makes it easy to build your own Deep Agents!

hwchase17's tweet image. 🧠🤖Deep Agents

Simple tool calling loops fail on longer time horizon or more complex tasks

Deep Agents like Deep Research, Claude Code & Manus succeed by using a number of tools and tricks

We created a new Python package which makes it easy to build your own Deep Agents!

Harrison Chase reposted

✈️ 🤖 AI Travel Agent An intelligent travel assistant that integrates real-time weather, search, and travel information. Leveraging multiple APIs, it streamlines everything from weather updates to currency conversion. Explore the guide on GitHub github.com/ashumishra2104…

LangChainAI's tweet image. ✈️ 🤖 AI Travel Agent

An intelligent travel assistant that integrates real-time weather, search, and travel information. Leveraging multiple APIs, it streamlines everything from weather updates to currency conversion.

Explore the guide on GitHub
github.com/ashumishra2104…

Harrison Chase reposted

🌐⚡ BLAST: AI Web Browser Engine A high-performance serving engine that adds web browsing to AI applications. BLAST provides an OpenAI-compatible interface with automatic parallelization, intelligent caching, and real-time streaming support. Explore this open-source project 👉…

LangChainAI's tweet image. 🌐⚡ BLAST: AI Web Browser Engine

A high-performance serving engine that adds web browsing to AI applications. BLAST provides an OpenAI-compatible interface with automatic parallelization, intelligent caching, and real-time streaming support.

Explore this open-source project 👉…

Harrison Chase reposted

🔍🤖 GraphQA Transform graph analysis into natural conversations with this NetworkX and LangChain-powered framework. Ask questions in plain English and let GraphQA select and execute the right algorithms, handling graphs up to 100K+ nodes. Check it out 👉…

LangChainAI's tweet image. 🔍🤖 GraphQA

Transform graph analysis into natural conversations with this NetworkX and LangChain-powered framework. Ask questions in plain English and let GraphQA select and execute the right algorithms, handling graphs up to 100K+ nodes.

Check it out 👉…

Harrison Chase reposted

@LangChain V1 middleware explained in < 30 seconds: 🔍 before_agent — Load files, validate input ✂️ before_model — Summarize convos, trim messages 🔄 wrap_model_call — Dynamic prompts, model, tools 🛠️ wrap_tool_call — Tool retries, error handling 🤵after_model — Human in the…

sydneyrunkle's tweet image. @LangChain V1 middleware explained in &amp;lt; 30 seconds:

🔍 before_agent — Load files, validate input
✂️ before_model — Summarize convos, trim messages
🔄 wrap_model_call — Dynamic prompts, model, tools
🛠️ wrap_tool_call — Tool retries, error handling
🤵after_model — Human in the…

Harrison Chase reposted

Just finished up a new guide to LangChain V1, with a comprehensive migration guide! Use this to test out our new create_agent primitive, which is highly customizable with agent middleware! Add human in the loop, summarization, and guardrails with just a few lines of code.…


Harrison Chase reposted

this highlights the issue with drag and drop build tools. it’s too limited for engineers and it’s too challenging to use for others.

I had to use dify in a project. Stuff that can be done in minutes in python take hours to days to implement in those workflow tools.



Harrison Chase reposted

🧠 Last Chance to Join: Context Engineering with LangChain & @ManusAI_HQ This webinar brings together @RLanceMartin Martin (Founding Engineer, LangChain) and Yichao “Peak” Ji (Co-founder & Chief Scientist, Manus; MIT Innovator Under 35) to share practical best practices in…


Harrison Chase reposted

You can now write custom code evals in JavaScript in addition to Python, giving you more flexibility and speed in your evaluation workflows. With JS code evals, get: ⚡️ Faster execution to reduce evaluation time and speed up your development 🔧 Work in your preferred stack:…


Harrison Chase reposted

We're hosting meetups in San Francisco, Boston, and NYC to celebrate LangChain's 3rd birthday and share some amazing news! Come hang with our engineering team, get a first look at our newest launches, and connect with other builders shipping agents in production. Plus: food,…

LangChainAI's tweet image. We&apos;re hosting meetups in San Francisco, Boston, and NYC to celebrate LangChain&apos;s 3rd birthday and share some amazing news!

Come hang with our engineering team, get a first look at our newest launches, and connect with other builders shipping agents in production. 

Plus: food,…
LangChainAI's tweet image. We&apos;re hosting meetups in San Francisco, Boston, and NYC to celebrate LangChain&apos;s 3rd birthday and share some amazing news!

Come hang with our engineering team, get a first look at our newest launches, and connect with other builders shipping agents in production. 

Plus: food,…
LangChainAI's tweet image. We&apos;re hosting meetups in San Francisco, Boston, and NYC to celebrate LangChain&apos;s 3rd birthday and share some amazing news!

Come hang with our engineering team, get a first look at our newest launches, and connect with other builders shipping agents in production. 

Plus: food,…

Harrison Chase reposted

I was demoing how our marketing twins collaborate with each other and watched the SEO specialist run for 3 min straight while using multiple APIs and MCPs. It scraped two competitor pages, ran keyword clustering, pulled SERP data, analyzed PASF patterns, mapped user intent…


a great event for our european friends! Intercom has done a lot of cool stuff with AI (Fin, etc)

Join us in Amsterdam for a fireside chat with Fergal Reid, Chief AI Officer at @intercom, as he shares how they built @Fin_ai with LangGraph —their AI agent that's transforming customer service. 📍Amsterdam | Thursday, October 9 at 6:00 pm Register → luma.com/FinxLangChain



Harrison Chase reposted

Join us in Amsterdam for a fireside chat with Fergal Reid, Chief AI Officer at @intercom, as he shares how they built @Fin_ai with LangGraph —their AI agent that's transforming customer service. 📍Amsterdam | Thursday, October 9 at 6:00 pm Register → luma.com/FinxLangChain


Harrison Chase reposted

.@clay's CEO Kareem Amin and Julia Schottenstein (LangChain's Head of GTM and Ops) break down the real product decisions behind Claygent — an AI research agent now running billions of queries for GTM teams. They cover why session replay shipped as P0, how to design UX when AI…

LangChainAI's tweet image. .@clay&apos;s CEO Kareem Amin and Julia Schottenstein (LangChain&apos;s Head of GTM and Ops) break down the real product decisions behind Claygent — an AI research agent now running billions of queries for GTM teams. They cover why session replay shipped as P0, how to design UX when AI…

Harrison Chase reposted

Tried the LangChain and LangGraph v1.0 Alpha yet? 👀It’s been a month since the release— and we’re calling on more devs to test it out + share feedback. 🧠New middleware API for agents 🧩Standard output/content blocks 📦New package + major API updates Adopt the alpha + tell us…


Harrison Chase reposted

not sure why langgraph doesn’t seem to get much love, I like more their code first approach and visualise what the agent does with state management and time travel, etc


Harrison Chase reposted

Great post from @hwchase17 about why @LangChainAI has *not* built their own workflow builder: blog.langchain.com/not-another-wo…


Harrison Chase reposted

Most people don't have the kind of structured thinking needed to build a useful agent. And the few who do (usually engineers) prefer more control than what a visual editor offers. Still useful, but the audience ends up being quite small.


Harrison Chase reposted

Responding to @harrisonchase's "Not Another Workflow Builder" with a twist: English IS the new programming language (shout out @karpathy ). In Austin, we're not typing—we're SPEAKING to Claude Code & Cursor to buildcolin agents. The interface layer is now conversational. 🗣️→🤖…


Harrison Chase reposted

“workflow builders” automate the obvious. Complex cases need thinking frameworks, not drag-and-drop boxes. SOPs + well-crafted prompts = scalable judgment, not just automation.

I am not excited about visual workflow builders 1. Not simple enough for the average user. I believe there should be a simpler way to create, modify, and adapt no-code agents 2. Not scalable for complex use cases Wrote a little blog: blog.langchain.com/not-another-wo…

hwchase17's tweet image. I am not excited about visual workflow builders

1. Not simple enough for the average user. I believe there should be a simpler way to create, modify, and adapt no-code agents

2. Not scalable for complex use cases

Wrote a little blog: blog.langchain.com/not-another-wo…
hwchase17's tweet image. I am not excited about visual workflow builders

1. Not simple enough for the average user. I believe there should be a simpler way to create, modify, and adapt no-code agents

2. Not scalable for complex use cases

Wrote a little blog: blog.langchain.com/not-another-wo…
hwchase17's tweet image. I am not excited about visual workflow builders

1. Not simple enough for the average user. I believe there should be a simpler way to create, modify, and adapt no-code agents

2. Not scalable for complex use cases

Wrote a little blog: blog.langchain.com/not-another-wo…
hwchase17's tweet image. I am not excited about visual workflow builders

1. Not simple enough for the average user. I believe there should be a simpler way to create, modify, and adapt no-code agents

2. Not scalable for complex use cases

Wrote a little blog: blog.langchain.com/not-another-wo…


Loading...

Something went wrong.


Something went wrong.