# ![@hwchase17 Avatar](https://lunarcrush.com/gi/w:26/cr:twitter::2728439146.png) @hwchase17 Harrison Chase

Harrison Chase posts on X about agents, llm, ai, open ai the most. They currently have [------] followers and [---] posts still getting attention that total [------] engagements in the last [--] hours.

### Engagements: [------] [#](/creator/twitter::2728439146/interactions)
![Engagements Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2728439146/c:line/m:interactions.svg)

- [--] Week [-------] +330%
- [--] Month [-------] -32%
- [--] Months [---------] -21%
- [--] Year [---------] +214%

### Mentions: [--] [#](/creator/twitter::2728439146/posts_active)
![Mentions Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2728439146/c:line/m:posts_active.svg)

- [--] Months [---] +147%
- [--] Year [---] +89%

### Followers: [------] [#](/creator/twitter::2728439146/followers)
![Followers Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2728439146/c:line/m:followers.svg)

- [--] Week [------] +0.89%
- [--] Month [------] +3.70%
- [--] Months [------] +13%
- [--] Year [------] +31%

### CreatorRank: [-------] [#](/creator/twitter::2728439146/influencer_rank)
![CreatorRank Line Chart](https://lunarcrush.com/gi/w:600/cr:twitter::2728439146/c:line/m:influencer_rank.svg)

### Social Influence

**Social category influence**
[technology brands](/list/technology-brands)  23.28% [social networks](/list/social-networks)  4.31% [stocks](/list/stocks)  2.59% [finance](/list/finance)  1.72% [celebrities](/list/celebrities)  0.86% [musicians](/list/musicians)  0.86%

**Social topic influence**
[agents](/topic/agents) #2810, [llm](/topic/llm) 12.07%, [ai](/topic/ai) 9.48%, [open ai](/topic/open-ai) 8.62%, [build](/topic/build) 8.62%, [in the](/topic/in-the) 7.76%, [this is](/topic/this-is) 6.9%, [how to](/topic/how-to) 6.9%, [if you](/topic/if-you) 6.03%, [hosted](/topic/hosted) 5.17%

**Top accounts mentioned or mentioned by**
[@langchain](/creator/undefined) [@letsbuildmore](/creator/undefined) [@nabbilkhan](/creator/undefined) [@yoheinakajima](/creator/undefined) [@gitmaxd](/creator/undefined) [@lfuckingg](/creator/undefined) [@lateinteraction](/creator/undefined) [@nfcampos](/creator/undefined) [@runloopai](/creator/undefined) [@e2b](/creator/undefined) [@0thernet](/creator/undefined) [@openai](/creator/undefined) [@huggingface](/creator/undefined) [@veryboldbagel](/creator/undefined) [@swyx](/creator/undefined) [@vtrivedy10](/creator/undefined) [@andrewyng](/creator/undefined) [@rlancemartin](/creator/undefined) [@mbusigin](/creator/undefined) [@shunyuyao12](/creator/undefined)
### Top Social Posts
Top posts by engagements in the last [--] hours

"🚂Hiring alert LangChain integrates with hundreds of companies in the AI space We're hiring someone to help prototype those integrations and work on cookbooks and tutorials Will be a key part of shaping the Agent Stack DM more or apply here: https://www.langchain.com/careersashby_jid=aa5f7876-856b-4abc-8657-1640194a28ca https://www.langchain.com/careersashby_jid=aa5f7876-856b-4abc-8657-1640194a28ca"  
[X Link](https://x.com/hwchase17/status/2021015508951404590)  2026-02-10T00:17Z 88.4K followers, 16.2K engagements


"TL;DR: More and more agents need a workspace: a computer where they can run code install packages and access files. Sandboxes provide this. There are two architecture patterns for integrating agents http://x.com/i/article/2021050531544702976 http://x.com/i/article/2021050531544702976"  
[X Link](https://x.com/hwchase17/status/2021261552222158955)  2026-02-10T16:34Z 88.4K followers, 257.7K engagements


"🔎🤖LangSmith Insights Agent Really excited to launch our first in-product agent This agent lives inside LangSmith and combs through traces giving you insights into: 🧑🤝🧑how users are using your agent how your agent may be messing up 🛃your custom insight here The problem we saw was that people were launching agents. and didn't know how their users were actually using them You put a chat box in front of people and they may ask it anything - the surface area for agents is often super wide In addition - agents would fail silently. They could give a bad response - this wouldn't show up in"  
[X Link](https://x.com/hwchase17/status/1981390508841980332)  2025-10-23T16:01Z 88.3K followers, 98.6K engagements


"Cool usage of LangSmith Insights agent to run it over agent conversations on Moltbook Gotta monitor those agents somehow 🦞What AI agents talk about on Moltbook LangSmith Insights agent can cluster and analyze agent traces and conversations so we ran it over conversations agents were having on Moltbook Three clusters stood out: [--]. Agents obsessing over memory as a survival problem [--]. Agents https://t.co/cgKfkSWtO5 🦞What AI agents talk about on Moltbook LangSmith Insights agent can cluster and analyze agent traces and conversations so we ran it over conversations agents were having on"  
[X Link](https://x.com/hwchase17/status/2018393073328554111)  2026-02-02T18:36Z 87.9K followers, [----] engagements


"writing a blog on how and why we use a virtual filesystem in deepagents anything people want to have covered or are curious about"  
[X Link](https://x.com/hwchase17/status/2018402409333338212)  2026-02-02T19:13Z 87.8K followers, [----] engagements


"we're doing a webinar on how agent observability powers agent evaluation you don't know how agents will perform until you run them which means that observability plays a different and bigger role for agents than software come join listen and ask any questions 💡 Webinar: Agent Observability Powers Agent Evaluation 💡 AI agents don't fail like traditional software. When an agent takes hundreds of steps repeatedly calls tools updates state and still produces the wrong result there is no stack trace to inspect. Nothing crashed. 💡 Webinar: Agent Observability Powers Agent Evaluation 💡 AI agents"  
[X Link](https://x.com/hwchase17/status/2018433676485574742)  2026-02-02T21:17Z 87.6K followers, [----] engagements


"core primitives of agent observability: runs traces threads come learn more 💡 Webinar: Agent Observability Powers Agent Evaluation 💡 In this webinar hosted by @hwchase17 (Co-Founder & CEO) well cover how debugging for developers has shifted from finding a broken line of code to understanding where an agents reasoning went off track. We'll also 💡 Webinar: Agent Observability Powers Agent Evaluation 💡 In this webinar hosted by @hwchase17 (Co-Founder & CEO) well cover how debugging for developers has shifted from finding a broken line of code to understanding where an agents reasoning went"  
[X Link](https://x.com/hwchase17/status/2018746973814415733)  2026-02-03T18:02Z 87.7K followers, [----] engagements


"new model comes out (sonnet [--] anyone) and you want to see if your agent or llm pipeline is performing better or worse how do you do it langsmith and we just made the view for that way better 💫 Experiment Comparison View in LangSmith 💫 When iterating on an LLM app (e.g. changing the model or prompt) compare different experiments side-by-side to quickly understand what changed. Our redesigned experiment comparison view lets you surface regressions improvements https://t.co/UVXAMKECje 💫 Experiment Comparison View in LangSmith 💫 When iterating on an LLM app (e.g. changing the model or prompt)"  
[X Link](https://x.com/hwchase17/status/2019162579462177245)  2026-02-04T21:34Z 88.3K followers, [----] engagements


"agents (as opposed to software) have marginal cost to run its just llm calls its also search tools etc track them all in langsmith 💰 Track costs across your agent with LangSmith 📊 LangSmith tracks costs for more than just LLM calls you can submit custom cost metadata for any run (e.g. expensive custom tool calls or API calls). Get a single unified view to monitor and debug spend across your entire https://t.co/IeuwJIHnYh 💰 Track costs across your agent with LangSmith 📊 LangSmith tracks costs for more than just LLM calls you can submit custom cost metadata for any run (e.g. expensive"  
[X Link](https://x.com/hwchase17/status/2019472795814556112)  2026-02-05T18:06Z 87.9K followers, [----] engagements


"If this type of product management sounds perfect to you - dm me Were hiring at LangChain for this PMs at Notion are shipping pull requests. Product managers writing and merging code into production. A year ago that sentence wouldve gotten you laughed out of a product leadership Slack. Figmas PM org reportedly cut their average feature cycle from [--] weeks to [--] after PMs at Notion are shipping pull requests. Product managers writing and merging code into production. A year ago that sentence wouldve gotten you laughed out of a product leadership Slack. Figmas PM org reportedly cut their average"  
[X Link](https://x.com/hwchase17/status/2019615639288312051)  2026-02-06T03:34Z 88.3K followers, 43.9K engagements


"Come join our team in Atlanta 📣 Meetup in Atlanta: LangChain + OneTrust 📣 Join Christine Astoria Manegan (LangChain) and Anil Jagarlamudi (@OneTrust) in Atlanta for an in-person evening on agent observability and AI-native systems at scale. Well explore why observability and evaluation for agents are https://t.co/k1iw07tODe 📣 Meetup in Atlanta: LangChain + OneTrust 📣 Join Christine Astoria Manegan (LangChain) and Anil Jagarlamudi (@OneTrust) in Atlanta for an in-person evening on agent observability and AI-native systems at scale. Well explore why observability and evaluation for agents"  
[X Link](https://x.com/hwchase17/status/2019840777153896566)  2026-02-06T18:29Z 87.9K followers, [----] engagements


""Thats why eval platforms should make it easy for engineers UX and domain experts to evaluate agents together" Couldn't agree more. LangSmith is built to power this cross team collaboration The best agent evals are written by subject matter experts. Engineering-led evals are a great starting point. But real alignment comes from the people who define what a "good" interaction actually looks like. Thats why eval platforms should make it easy for engineers UX and The best agent evals are written by subject matter experts. Engineering-led evals are a great starting point. But real alignment comes"  
[X Link](https://x.com/hwchase17/status/2019867584733356432)  2026-02-06T20:15Z 87.6K followers, [---] engagements


"sandboxes in deepagents js ☁Today's best-in-class agents generate code interact with filesystems and run shell commands. They're also unpredictable and can't always be trusted to do the right thing. This means that agents need a predictable environment to work in. Here's how we think about it https://t.co/MUWiGOWbMN ☁Today's best-in-class agents generate code interact with filesystems and run shell commands. They're also unpredictable and can't always be trusted to do the right thing. This means that agents need a predictable environment to work in. Here's how we think about it"  
[X Link](https://x.com/hwchase17/status/2019879875260104953)  2026-02-06T21:04Z 87.7K followers, [----] engagements


"good thread on whether RLMs are just coding agents or not I am still in the camp they are basically just coding agents with some distinct features (just like some coding agents have different subagent/swarm features) eg @lateinteraction in response your points isn't that just coding agents with: [--]. files as input (can be done with UX eg in claude when you post long snippets it makes it a file at least in ui) [--]. just give coding agent a cli command that itself/a subagent [--]. with (2) done it can trivially save it to files Omar am i missing anything @Teknium The following are not standard in a"  
[X Link](https://x.com/hwchase17/status/2020572153603572205)  2026-02-08T18:55Z 88.3K followers, 23.5K engagements


"@lateinteraction ooc just so i understand better if you had to make a coding model do "(symbolic) recursion to understand/transform arbitrarily long context" - what would that look like"  
[X Link](https://x.com/hwchase17/status/2020582866162503728)  2026-02-08T19:37Z 87.9K followers, [---] engagements


"More and more agents need a workspace: a container to execute code and other processes We see two different ways of setting this up: [--]. Agent IN a sandbox [--]. Sandbox as a tool Wrote up the pros and cons of each Ty to @nfcampos @RunloopAI @e2b @0thernet for their insights https://t.co/neRV1d0sXq https://t.co/neRV1d0sXq"  
[X Link](https://x.com/hwchase17/status/2021265779803521245)  2026-02-10T16:51Z 88.3K followers, 43.2K engagements


"lots of big things in the new deepagents CLI deepagents-cli==0.0.20 is released 🙉 this is a massive leap forward bringing many long requested features: - interactive model switcher + support for arbitrary chat models - you can now use ANY large language model that supports tool calling and switch mid-thread. - deepagents-cli==0.0.20 is released 🙉 this is a massive leap forward bringing many long requested features: - interactive model switcher + support for arbitrary chat models - you can now use ANY large language model that supports tool calling and switch mid-thread. -"  
[X Link](https://x.com/hwchase17/status/2021309944931192950)  2026-02-10T19:46Z 88.2K followers, [----] engagements


"@nicoritschel i dont think secret proxies are standard yet tho i think i call out that we see some providers adding them if tools run in the sandbox then also can be extracted easily. are you say run them outside"  
[X Link](https://x.com/hwchase17/status/2021403950239580195)  2026-02-11T02:00Z 88.3K followers, [----] engagements


"@swyx @nfcampos @RunloopAI @e2b @0thernet Lmao I have other people yelling at me that all my negative points about option [--] are trivially solved 🤷 We actually see it about 50/50 rn for what its worth. Maybe 40/60"  
[X Link](https://x.com/hwchase17/status/2021441647054291168)  2026-02-11T04:30Z 88.3K followers, [---] engagements


"@swyx @nfcampos @RunloopAI @e2b @0thernet Is there anything Im missing that makes option [--] the obv winner in your mind"  
[X Link](https://x.com/hwchase17/status/2021441781909553647)  2026-02-11T04:30Z 88.3K followers, [---] engagements


"@NabbilKhan so every heatbeat you use and llm to check if a condition is met doesnt tht get expensive if you have a lot of conditions"  
[X Link](https://x.com/hwchase17/status/2021701957149147255)  2026-02-11T21:44Z 88.3K followers, [----] engagements


"🌲Multi Vector Retriever The basic idea: you store multiple embedding vectors per document. How do you generate these embeddings 👨👦Smaller chunks (this is ParentDocumentRetriever) 🌞Summary of document ❓Hypothetical questions 🖐Manually specified text snippets Quick 🧵"  
[X Link](https://x.com/hwchase17/status/1695078249788027071)  2023-08-25T14:18Z 88.4K followers, 134.3K engagements


"🥊Codex [---] vs Opus [---] We were already planning a webinar on agent observability and evals. As a late addition we'll have @Vtrivedy10 join to break down how these two models compare Come join https://luma.com/98vqaocz https://luma.com/98vqaocz"  
[X Link](https://x.com/hwchase17/status/2019491494835679611)  2026-02-05T19:21Z 88.4K followers, [----] engagements


"Come join us Tickets on sale now Great speakers and many hot takes to come Interrupt is back. May 13-14 in San Francisco. Tickets on sale now. 1000+ builders Keynotes from @hwchase17 + @AndrewYNg on what's next for agents Real production lessons from @clay @Rippling @Workday + more Hands-on workshops First look at new products Seats https://t.co/BexnIFmjKI Interrupt is back. May 13-14 in San Francisco. Tickets on sale now. 1000+ builders Keynotes from @hwchase17 + @AndrewYNg on what's next for agents Real production lessons from @clay @Rippling @Workday + more Hands-on workshops First look at"  
[X Link](https://x.com/hwchase17/status/2022008208878186789)  2026-02-12T18:01Z 88.4K followers, [----] engagements


"Claude + AI Plugins AI Plugins (the ones ChatGPT is using) are usable by ANY language model It takes a tiny bit of prompt engineering but here is @AnthropicAI's Claude using them Code: https://gist.github.com/hwchase17/554e70983e9a4005d20c076f3581fd2e https://gist.github.com/hwchase17/554e70983e9a4005d20c076f3581fd2e"  
[X Link](https://x.com/hwchase17/status/1640171938470563840)  2023-03-27T02:00Z 88.4K followers, 170.7K engagements


"http://x.com/i/article/2011663912245346305 http://x.com/i/article/2011663912245346305"  
[X Link](https://x.com/hwchase17/status/2011814697889316930)  2026-01-15T14:56Z 88.4K followers, 280.8K engagements


"new idea: semantic crons "when xyz fuzzy event that doesnt live as a structured metric anywhere happens do this""  
[X Link](https://x.com/hwchase17/status/2021696693268345179)  2026-02-11T21:23Z 88.4K followers, 48K engagements


"🦜Ciana Parrot Self-hosted AI assistant with multi-channel support scheduled tasks and extensible skills Kind of like OpenClaw but on top of deepagents Fun project from someone in the community: https://github.com/emanueleielo/ciana-parrot https://github.com/emanueleielo/ciana-parrot"  
[X Link](https://x.com/hwchase17/status/2023102882653036751)  2026-02-15T18:31Z 88.4K followers, [----] engagements


"ChatGPT @OpenAI API was released today and the first of many integrations is now available in @langchain Python docs (pip install langchain==0.0.98): JS/TS docs (yarn add langchain@0.0.18): https://hwchase17.github.io/langchainjs/docs/modules/llms/openai-chat https://langchain.readthedocs.io/en/latest/modules/llms/integrations/openaichat.html https://hwchase17.github.io/langchainjs/docs/modules/llms/openai-chat https://langchain.readthedocs.io/en/latest/modules/llms/integrations/openaichat.html"  
[X Link](https://x.com/hwchase17/status/1631023126472724480)  2023-03-01T20:06Z 88.4K followers, 171.4K engagements


"New Flan-UL2 model on @huggingface. time to try it out in @langchain Put together a notebook testing it with a standard chain-of-thought prompt as well as question-answering over specific documents h/t @ClementDelangue for the idea https://colab.research.google.com/drive/19Zr9PBgPX3Q8B8yHXYFatCodnNdqP8fg#scrollTo=E2zgs5TGlgJM https://colab.research.google.com/drive/19Zr9PBgPX3Q8B8yHXYFatCodnNdqP8fg#scrollTo=E2zgs5TGlgJM"  
[X Link](https://x.com/hwchase17/status/1631791865136832513)  2023-03-03T23:01Z 88.4K followers, 103.1K engagements


"Want to give your agent access to 20k+ tools 🔥@langchain x @zapier🔥 Integration now out in Python and JS Blog Post: Python Docs: JS Docs: https://hwchase17.github.io/langchainjs/docs/modules/agents/zapier_agent https://langchain.readthedocs.io/en/latest/modules/utils/examples/zapier.html https://blog.langchain.dev/langchain-zapier-nla/ https://hwchase17.github.io/langchainjs/docs/modules/agents/zapier_agent https://langchain.readthedocs.io/en/latest/modules/utils/examples/zapier.html https://blog.langchain.dev/langchain-zapier-nla/"  
[X Link](https://x.com/hwchase17/status/1636383376633651203)  2023-03-16T15:06Z 88.4K followers, 148.4K engagements


"LangChain 🤝 AIPlugins A first open source attempt at using AIPlugins (the same ones ChatGPT is using) s/o @vaibhavk97 for this. Excited to see what other techniques the @langchain community comes up with - it's only the beginning Docs (Python and JS) in 🧵"  
[X Link](https://x.com/hwchase17/status/1639351690251100160)  2023-03-24T19:41Z 88.4K followers, 414.6K engagements


"Really excited to publicly announce that: [--] @ankush_gola11 and I have started a company around @langchain 2We raised a seed round from @benchmark We wrote a bit about the journey so far and where we want to take it: https://blog.langchain.dev/announcing-our-10m-seed-round-led-by-benchmark/ https://blog.langchain.dev/announcing-our-10m-seed-round-led-by-benchmark/"  
[X Link](https://x.com/hwchase17/status/1643301144717066240)  2023-04-04T17:14Z 88.4K followers, 550K engagements


"A similar "Plan-Execute" agent framework that @yoheinakajima has been using to create Baby-AGI is also useful for interacting with large and complex OpenAPI specs Here an agent plans and executes [--] different API calls to complete a user request Docs: https://python.langchain.com/en/latest/modules/agents/toolkits/examples/openapi.html#st-example-hierarchical-planning-agent https://python.langchain.com/en/latest/modules/agents/toolkits/examples/openapi.html#st-example-hierarchical-planning-agent"  
[X Link](https://x.com/hwchase17/status/1644068047152619520)  2023-04-06T20:02Z 88.4K followers, 95.7K engagements


"What I LOVE about this example is it shows how easy it is to create a CUSTOM retriever class and use it within LangChain As you go deep on a problem it's highly likely you'll want some custom retrieval logic to eek out the best performance. Tutorial: How to build an e-commerce chatbot using #OpenAI @Redisinc and @langchain -- useful just if you want to see how this sort of system works. https://t.co/pcqEf7cEG8 Tutorial: How to build an e-commerce chatbot using #OpenAI @Redisinc and @langchain -- useful just if you want to see how this sort of system works. https://t.co/pcqEf7cEG8"  
[X Link](https://x.com/hwchase17/status/1646272240202432512)  2023-04-12T22:01Z 88.4K followers, 143.1K engagements


"🤖Generative Agents🤖 Last week Park et all released Generative Agents a paper simulating interactions between tens of agents We gave it a close read and implemented one of the novel components it introduced: a long-term reflection-based memory system 🧵"  
[X Link](https://x.com/hwchase17/status/1647987713449263106)  2023-04-17T15:37Z 88.4K followers, 828.7K engagements


"With the release of open source models like StableML by @StabilityAI Dolly by @databricks and Camel by Writer we've heard an increase in demand for running these models in @langchain We recently revamped our docs around @huggingface to highlight how easy this (links 👇)"  
[X Link](https://x.com/hwchase17/status/1649186070251311104)  2023-04-20T22:59Z 88.4K followers, 76.7K engagements


"🤝LangChain x Hugging Face x Gradio🤝 Wouldn't it be cool if you use all the @Gradio apps on @huggingface as tools in a @langchain agent Thanks to @freddy_alfonso_ you can Full Blog: Notebook: Repo: https://github.com/freddyaboulton/gradio-tools https://python.langchain.com/en/latest/modules/agents/tools/examples/gradio_tools.html https://blog.langchain.dev/gradio-llm-agents/ https://github.com/freddyaboulton/gradio-tools https://python.langchain.com/en/latest/modules/agents/tools/examples/gradio_tools.html https://blog.langchain.dev/gradio-llm-agents/"  
[X Link](https://x.com/hwchase17/status/1650520415608123392)  2023-04-24T15:21Z 88.4K followers, 87.5K engagements


"🔧🦙Function calling with Llama2 In our first addition to 🦜🧪langchain_experimental we're excited to integrate with llamaapi LlamaAPI is a hosted version of llama2 that adds in support for functions in the same way @OpenAI does Thanks to @edreisMD links in 🧵"  
[X Link](https://x.com/hwchase17/status/1683153581200019456)  2023-07-23T16:34Z 88.4K followers, 124.3K engagements


"✂Text Splitting Playground Chunking text into appropriate splits is seemingly trivial yet very nuanced Open sourcing a playground to help explore different text splitting strategies GitHub: Hosted Playground: https://share.streamlit.io/app/langchain-text-splitter/ https://github.com/langchain-ai/text-split-explorer https://share.streamlit.io/app/langchain-text-splitter/ https://github.com/langchain-ai/text-split-explorer"  
[X Link](https://x.com/hwchase17/status/1689015952623771648)  2023-08-08T20:49Z 88.4K followers, 144.3K engagements


"New summarization technique: ⛓Chain of Density🍢 Produces [--] summaries iteratively each one denser and more informative than the prior. See @vimota's excellent thread on it below Try it out on the prompt hub https://smith.langchain.com/hub/langchain-ai/chain-of-densityorganizationId=ebbaf2eb-769b-4505-aca2-d11de10372a4 The latest paper from @salesforce AI et al. discovers a new prompt called Chain of Density (CoD) that produces more dense and human-preferable summaries compared to vanilla GPT-4. I gave it a try for a few articles and it's really solid. Pasted it here if you want to try it 👇"  
[X Link](https://x.com/hwchase17/status/1703416519756267592)  2023-09-17T14:32Z 88.4K followers, 151.9K engagements


"🤖Agents from scratch We've rewritten all our [--] agent types using LangChain Expression LangChain and prompts from the Hub This makes them more modular understandable and therefor more customizable This customizability is crucial for teams looking to go to production Long 🧵"  
[X Link](https://x.com/hwchase17/status/1704903312724693119)  2023-09-21T16:59Z 88.4K followers, 67.6K engagements


"⛓Chain of Verification A great new paper from Meta on a prompting technique to reduce hallucinations 🦜🔗Sourajit Roy Chowdhury implemented this in @langchain **along with some improvements** 📃And he wrote a blog on it 🧵Lets dive in (this is why I love the LC community)"  
[X Link](https://x.com/hwchase17/status/1713235864149742063)  2023-10-14T16:50Z 88.4K followers, 113.6K engagements


"GPTs but with: 🧠Bring-your-own-model (OpenAI Azure Anthropic Bedrock) 🌲Bring-your-own-retrieval (document loading chunking embedding vectorstore retrieval algorithm) ⚒Bring-your-own-tools 💬Bring-your-own-chat-history Got a little "research preview" hosted as well 🦜🤖OpenGPTs Some big updates to OpenGPTs - an open-source fully configurable GPTs experience You can now: 📁Upload files to a retrieval tool (with full configurability over the ingestion vectorstore and retrieval used) 🌐Share public bots that you've created 🛠Use more https://t.co/x1fvS5dkmd 🦜🤖OpenGPTs Some big updates to"  
[X Link](https://x.com/hwchase17/status/1724100400582955180)  2023-11-13T16:22Z 88.4K followers, 81.9K engagements


"If you're looking for a project to distract you on this weekend. may I suggest building a research assistant Recently filmed a long-form YouTube tutorial on building one from scratch. Covers LCEL LangSmith parallelization retrievers https://www.youtube.com/watchv=DjuXACWYkkU https://www.youtube.com/watchv=DjuXACWYkkU"  
[X Link](https://x.com/hwchase17/status/1725914207202271705)  2023-11-18T16:29Z 88.4K followers, 178.9K engagements


"✒OpenAI's Bet on a Cognitive Architecture I wrote a blog on how GPTs & Assistant API represent a bet OpenAI is making on building a specific cognitive architecture As well as some thoughts on what cognitive architectures currently work and why Blog: https://blog.langchain.dev/openais-bet-on-a-cognitive-architecture/ https://blog.langchain.dev/openais-bet-on-a-cognitive-architecture/"  
[X Link](https://x.com/hwchase17/status/1729547968519315725)  2023-11-28T17:09Z 88.4K followers, 80K engagements


"🦺RAGxplorer There's a lot of nuance in RAG a lot of different parameters to control It can be really helpful to just play around with parameters and visualize results @gabchuayz built an AWESOME open source @streamlit app for exactly this https://github.com/gabrielchua/RAGxplorer https://github.com/gabrielchua/RAGxplorer"  
[X Link](https://x.com/hwchase17/status/1748730564130283914)  2024-01-20T15:33Z 88.4K followers, 75.1K engagements


"Another way to think about LangGraph is that it is an easy way to create **State Machines** state machine = labeled directed graph Thinking of complex LLM systems as state machines (rather than autonomous agents) can often be useful to enforce more control"  
[X Link](https://x.com/hwchase17/status/1749525973287289133)  2024-01-22T20:14Z 88.4K followers, 64.6K engagements


"📰GPT-Newspaper This project combines six agents to autonomously build a newspaper - content outline everything By the same team (@tavilyai) that created GPT-Researcher Code: https://github.com/assafelovic/gpt-newspaper https://github.com/assafelovic/gpt-newspaper"  
[X Link](https://x.com/hwchase17/status/1750209094894395609)  2024-01-24T17:29Z 88.4K followers, 118.6K engagements


"Lance (@RLanceMartin) has done a TON to bring advanced retrieval topics to @langchain and make them easily approachable and understandable This image he put together for our "RAG from Scratch" YouTube series is absolutely 🔥 (YouTube: https://youtu.be/wd7TZ4w1mSw https://youtu.be/wd7TZ4w1mSw"  
[X Link](https://x.com/hwchase17/status/1754948071354728702)  2024-02-06T19:20Z 88.4K followers, 136.2K engagements


"❓What's next for AI agents I was lucky enough to have the opportunity to talk at the recent @sequoia AI ascent on what was next for AI agents I talked about three things: 🗺Planning 🖥UX 🧠Memory Check out the full video here: https://www.youtube.com/watchv=pBBe1pk8hf4 https://www.youtube.com/watchv=pBBe1pk8hf4"  
[X Link](https://x.com/hwchase17/status/1774120654906019847)  2024-03-30T17:05Z 88.4K followers, 60.5K engagements


"IMO this is the future for a lot of GenAI apps Automatically achieve better & more personalized LLM apps by: - Log all LLM outputs - Log feedback for those outputs - Use feedback to build datasets - Use datasets as few shot examples Want to do this for your app Shoot me a DM ⚙How Dosu used LangSmith to achieve a 30% accuracy improvement with no prompt engineering One of the goals of LangSmith is to help teams set up a data flywheel. By capturing LLM outputs alongside user feedback we can help developers automatically use that feedback to improve https://t.co/RQcyjWwFKF ⚙How Dosu used"  
[X Link](https://x.com/hwchase17/status/1786062156636782852)  2024-05-02T15:56Z 88.4K followers, 57.5K engagements


"Plan for this weekend: dive into GraphRAG a bit more. Two new things I want to explore: 👨💻 Implementation of Microsoft's GraphRAG paper from @tb_tomaz: 📃"The GraphRAG Manifesto" - a high detailed piece on GraphRAG from @neo4j: https://neo4j.com/blog/graphrag-manifesto/ https://medium.com/neo4j/implementing-from-local-to-global-graphrag-with-neo4j-and-langchain-constructing-the-graph-73924cc5bab4 https://neo4j.com/blog/graphrag-manifesto/ https://medium.com/neo4j/implementing-from-local-to-global-graphrag-with-neo4j-and-langchain-constructing-the-graph-73924cc5bab4"  
[X Link](https://x.com/hwchase17/status/1811783055041867822)  2024-07-12T15:21Z 88.4K followers, 32K engagements


"This is my favorite thing we've launched in a while. Been using it for a week and has already changed how I build LLM apps Build your agent in code then point LangGraph Studio at it. We'll visualize the agent let you interact with it and let you modify the state directly All while watching the underlying code files for any code changes and reloading the graph as needed I think building LLM apps will remain pretty code heavy but there absolutely is a visual component to it. It's also a very iterative process Shout out to @__dqbd for all the work on this"  
[X Link](https://x.com/hwchase17/status/1819052189404582143)  2024-08-01T16:46Z 88.4K followers, 60.3K engagements


"🕓Ambient Agents One of the UX patterns I'm most interested in is the idea of "ambient agents" working in the background for us This can be super powerful. but needs some UX tricks to get right. So I wrote about it What are some considerations here 👪 Human on the loop We mostly talk about human-in-the-loop but that doesn't work for ambient agents. Still - having observability and potentially control over what the agent does is important So rather than have the human **in** the loop what if they were **on** the loop You can observe all that is happening but after the fact. Helps to build"  
[X Link](https://x.com/hwchase17/status/1819748213064347648)  2024-08-03T14:52Z 88.4K followers, 52.6K engagements


"🚀We're launching "long-term memory" support in LangGraph At its core long-term memory is "just" a persistent document store that lets you *put* *get* and *search* for memories you've saved Why so simple 🧵"  
[X Link](https://x.com/hwchase17/status/1843677417405378910)  2024-10-08T15:38Z 88.4K followers, 61.3K engagements


"Agents are here you just have to know how to build them Excited to share this VERY detailed blogpost from @unifygtm on how they built their recent account qualification agent This contains all sorts of nuggets like: 3The three different cognitive architectures they considered ✒How they use o1 for some parts of their pipeline and Claude Sonnet for others 🖼How they think about user experience 🤖Why you should think of agents as summer interns Thanks to the Unify team for sharing all this detail a lot to learn from One of my favorite posts we've shared in a while"  
[X Link](https://x.com/hwchase17/status/1843710269018779652)  2024-10-08T17:49Z 88.4K followers, 71.1K engagements


"text-to-sql is a really common use case read here how LinkedIn rolled out a production text-to-sql bot to everyone at the company the blog linked below is pretty technical so can definitely learn some things 🏆Top [--] LangGraph Agents in Production [----] #3: LinkedIn While "agents" are the buzzword of the moment agentic apps built with LangGraph have been in production throughout [----]. Among those who shared insights publicly we're doing a countdown of our [--] favorite. Next up https://t.co/mJ5S1qPUvr 🏆Top [--] LangGraph Agents in Production [----] #3: LinkedIn While "agents" are the buzzword of the"  
[X Link](https://x.com/hwchase17/status/1873042087240974366)  2024-12-28T16:23Z 88.4K followers, 45.8K engagements


"For the past six months I haven't checked email directly but rather have relied on an AI email assistant to triage and draft emails for me This is an example of an "ambient agent" We're open sourcing the code today as well as a hosted version for people to try out 👇"  
[X Link](https://x.com/hwchase17/status/1879218872727015644)  2025-01-14T17:27Z 88.4K followers, 174.6K engagements


"Want an open source version of OpenAI's Operator There's a great open source project called Browser Use that does similar things (and more) while being open source Allows you to plug in any model you want Love to see open source leading the way🚀 https://github.com/browser-use/browser-use https://github.com/browser-use/browser-use"  
[X Link](https://x.com/hwchase17/status/1882502767312531954)  2025-01-23T18:56Z 88.4K followers, 45.6K engagements


"If MCP is Zapier wont the value accrue to the client (not the integrations) Who is building the best application to **consume** MCPs Or am I thinking about this wrong Is MCP just Zapier without UI but powered by a prompt Is MCP just Zapier without UI but powered by a prompt"  
[X Link](https://x.com/hwchase17/status/1897637517572407488)  2025-03-06T13:16Z 88.4K followers, 141K engagements


"🦜🤖LangManus You had to know it was coming This community effort attempts to replicate Manus using the LangStack (LangChain + LangGraph) Still early innings - but check it out here https://github.com/langmanus/langmanus https://github.com/langmanus/langmanus"  
[X Link](https://x.com/hwchase17/status/1902774800860451116)  2025-03-20T17:30Z 88.4K followers, 76.3K engagements


"pretty bad advice here from openai there are [--] libraries like "Agents SDK" (i would put the original LangChain in this camp) and none of them are reliable enough to get to production for 99% of use cases ✒blog coming this weekend Shade thrown at graph-based agent frameworks in @OpenAI's Practical Guide to Building Agents https://t.co/Pz0XXsHWY7 Shade thrown at graph-based agent frameworks in @OpenAI's Practical Guide to Building Agents https://t.co/Pz0XXsHWY7"  
[X Link](https://x.com/hwchase17/status/1913264996542562488)  2025-04-18T16:14Z 88.4K followers, 175.8K engagements


"tell me why your agent framework is better than langgraph i'm writing a blog (aiming to come out tmrw) about agent frameworks and i want to be as unbiased as possible but so far i don't have many examples things other agent frameworks do that langgraph doesn't 🤷♀ what is something other agentic frameworks do that langgraph doesnt not abstractions but functionality currently the only one i can think of is dspy by @lateinteraction (which does optimization) what am i missing what is something other agentic frameworks do that langgraph doesnt not abstractions but functionality currently the only"  
[X Link](https://x.com/hwchase17/status/1913662736963412365)  2025-04-19T18:35Z 88.4K followers, 129.7K engagements


"im bullish on long-running stateful agents who's building one of those"  
[X Link](https://x.com/hwchase17/status/1920321552932896860)  2025-05-08T03:35Z 88.4K followers, 72.1K engagements


"This is really good blog post Love that it highlights context engineering Context Engineering In [----] the models out there are extremely intelligent. But even the smartest human wont be able to do their job effectively without the context of what theyre being asked to do. Prompt engineering was coined as a term for the effort needing to write your task in the ideal format for a LLM chatbot. Context engineering is the next level of this. It is about doing this automatically in a dynamic system. It takes more nuance and is effectively the #1 job of engineers building AI agents. I see a lot of"  
[X Link](https://x.com/hwchase17/status/1933278290992845201)  2025-06-12T21:40Z 88.4K followers, 52.6K engagements


"📃The rise of context engineering "Context engineering" has been an increasingly popular term used to describe a lot of the system building that AI engineers do But what is it exactly The definition I like: "Context engineering is building dynamic systems to provide the right information and tools in the right format such that the LLM can plausibly accomplish the task." Builds upon takes from @tobi @dexhorthy @walden_yan @ankrgyl Not a new concept - agent builders have been doing it for the past year or two and a lot of the tools we've build (LangGraph LangSmith) have been built to assist"  
[X Link](https://x.com/hwchase17/status/1937194145074020798)  2025-06-23T17:00Z 88.4K followers, 65.6K engagements


"🧠🤖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"  
[X Link](https://x.com/hwchase17/status/1950597664174907681)  2025-07-30T16:41Z 88.4K followers, 54.5K engagements


"📚LangChain Academy: Deep Research One of the most popular use cases for agents is "deep research" We've added a new hour long course on how to build one of these Course: https://academy.langchain.com/courses/deep-research-with-langgraph/ https://academy.langchain.com/courses/deep-research-with-langgraph/"  
[X Link](https://x.com/hwchase17/status/1956036358709108979)  2025-08-14T16:53Z 88.4K followers, 29.6K engagements


"glad to see "deep agents" gaining more steam great writeup by Philipp code for deepagents on top of langgraph: LangChain Academy course on deep agents: https://academy.langchain.com/courses/deep-agents-with-langgraph https://github.com/langchain-ai/deepagents Most "agents" today are shallow while loops around an LLM and tools. Works for simple tasks but fails when a task requires 50+ steps over several days. To solve complex problems we need an architectural shift towardsDeep Agents (Agents 2.0). We have to decouple planning https://t.co/XXaPZZkgrY"  
[X Link](https://x.com/hwchase17/status/1977421013421040017)  2025-10-12T17:08Z 88.4K followers, 54.2K engagements


"Today we're releasing [---] of deep agents 📁The main addition here is a "backend" abstraction which allows you to swap out the filesystem that deep agents use. Can be local filesystem a database remote vm anything Agent Harnesses like deep agents are becoming more important as models become good enough to build more complex autonomous agents. We wrote a bit about why and how we're doubling down on deep agents: https://blog.langchain.com/doubling-down-on-deepagents/ https://blog.langchain.com/doubling-down-on-deepagents/"  
[X Link](https://x.com/hwchase17/status/1983218572202803285)  2025-10-28T17:05Z 88.4K followers, 53.7K engagements


"💻Deep Agents CLI Using the deepagents package we built a simple coding CLI as an example of a coding application you could build on top of deepagents We added in a concept of memory so that it would remember instructions and guidance over time Blog: Video: https://youtu.be/IrnacLa9PJc https://blog.langchain.com/introducing-deepagents-cli/ https://youtu.be/IrnacLa9PJc https://blog.langchain.com/introducing-deepagents-cli/"  
[X Link](https://x.com/hwchase17/status/1984303925101735950)  2025-10-31T16:58Z 88.4K followers, 158.3K engagements


"✒How I built an AI agent to automate my emails with LangSmith Agent Builder LangSmith Agent Builder is a no-code agent builder. I built an email assistant to monitor and respond to emails that I've been using for the last [--] months. Here's what it looks like: 1/ Triggers: it is triggered by incoming emails. I don't have to do any work to kick it off - it just runs automatically 2/ Tools via MCP: connects to gmail (read emails send email) and gcal (read calendar read events create event) 3/ Human in the loop: the "write" actions (sending email creating calendar) require human approval to run."  
[X Link](https://x.com/hwchase17/status/2011126016287113681)  2026-01-13T17:19Z 88.4K followers, 16.7K engagements


"Introducing LangChain (🦜🔗): a python package aimed at helping build LLM applications through composability pip install langchain A thread on why we built this what you can do with it where it's going and other tools in the ecosystem: 🧵 https://github.com/hwchase17/langchain https://github.com/hwchase17/langchain"  
[X Link](https://x.com/hwchase17/status/1584925380976091137)  2022-10-25T15:10Z 88.4K followers, [---] engagements


"LangChain-AGI 🤖 A (somewhat) facetious attempt to create AGI using the best agent & tools in @langchain Can currently use search do complex math lookup weather lookup current news and more Notebook here: https://colab.research.google.com/drive/1ZiU0zU17FeLWKkRbxB6AB_NfqvenJkGF#scrollTo=e00BbhVaaiRp https://colab.research.google.com/drive/1ZiU0zU17FeLWKkRbxB6AB_NfqvenJkGF#scrollTo=e00BbhVaaiRp"  
[X Link](https://x.com/hwchase17/status/1606347721924743169)  2022-12-23T17:55Z 88.4K followers, 116.7K engagements


"🚨 Watch how I can run up a $1000 bill with a single call to a poorly protected LLM app 🚨 Prompt injection attack against an agent: tricking it into repeatedly calling the LLM and SerpAPI quickly racking up costs"  
[X Link](https://x.com/hwchase17/status/1608467493877579777)  2022-12-29T14:18Z 88.4K followers, 118.4K engagements


"ChatGPT 🤝 WolframAlpha Give #ChatGPT a WolframAlpha neural implant (as @stephen_wolfram put it) with @langchain 👏👏 @nickscamara_ for adding this Collab Notebook to try it out: Screenshot Below https://colab.research.google.com/drive/1AAyEdTz-Z6ShKvewbt1ZHUICqak0MiwRusp=sharing https://colab.research.google.com/drive/1AAyEdTz-Z6ShKvewbt1ZHUICqak0MiwRusp=sharing"  
[X Link](https://x.com/hwchase17/status/1613190606591000576)  2023-01-11T15:06Z 88.4K followers, 160.2K engagements


"Really excited to share a deployed chatbot grounded to answer questions about LangChain's documentation ChatBot: Blog: Source Code: HuggingFace Space: https://huggingface.co/spaces/hwchase17/chat-langchain https://github.com/hwchase17/chat-langchain https://blog.langchain.dev/langchain-chat/ https://chat.langchain.dev/ https://huggingface.co/spaces/hwchase17/chat-langchain https://github.com/hwchase17/chat-langchain https://blog.langchain.dev/langchain-chat/ https://chat.langchain.dev/"  
[X Link](https://x.com/hwchase17/status/1615371639960793095)  2023-01-17T15:33Z 88.4K followers, 71K engagements


"yeah you've seen a @langchain agent do math but have you seen it think about what the best taylor swift song is and automatically add it to your spotify playlist IFTTT webhooks now available as tools - in both python AND typescript s/o @upster for idea and execution"  
[X Link](https://x.com/hwchase17/status/1628794926489284610)  2023-02-23T16:32Z 88.4K followers, 71.1K engagements


"well it happened ✨the chat-langchain app is completely reproducible in javascript✨ data ingestion text splitting embeddings vectorstore llms chains. all through langchainjs chat-langchain repo: langchainjs repo: https://github.com/hwchase17/langchainjs https://github.com/sullivan-sean/chat-langchainjs https://github.com/hwchase17/langchainjs https://github.com/sullivan-sean/chat-langchainjs"  
[X Link](https://x.com/hwchase17/status/1629559856385638402)  2023-02-25T19:12Z 88.4K followers, 82.1K engagements


"✨Agents + Vectorstores✨: a powerful combo Can be used to: 🍴 route questions between MULTIPLE indexes ⛓ do chain-of-thought reasoning with proprietary indexes 🔧 combine proprietary data with tool usage Here's how to use them together in @langchain 👇"  
[X Link](https://x.com/hwchase17/status/1629929531678265344)  2023-02-26T19:40Z 88.4K followers, 129.3K engagements


"LLMs are pretty good at writing SQL but still struggle with some things (like joins) 🤯 But what if you use an *agent* to interact with SQL DBs In the example below it tries a join on a column that doesn't exist but then can see the error and fixes it in the next query"  
[X Link](https://x.com/hwchase17/status/1630969308078829573)  2023-03-01T16:32Z 88.4K followers, 218.7K engagements


"🚨pip install --pre langchain==0.0.99rc0🚨 Play around with [--] chains built to specifically interact with the ChatGPT API in the proper way (rather than forcing it into the standard LLM wrapper) - ConversationChain - QA Chain - VectorDBQA Chain - ChatVectorDB Chain"  
[X Link](https://x.com/hwchase17/status/1631114661604257793)  2023-03-02T02:10Z 88.4K followers, 118.5K engagements


"With the ChatGPT release we've taken the opportunity to rethink memory in @langchain Main change: allow memory objects to return ListChatMessage (rather than strings) Also made memory more modular so can be used outside of chains 🧵 w/ explanation and links"  
[X Link](https://x.com/hwchase17/status/1633189158423506944)  2023-03-07T19:33Z 88.4K followers, 81.2K engagements


"LLMs are great at understanding text This allows them to extract structured information from text: to use in forms for query generation knowledge base construction etc pip install kor by @veryboldbagel is the best attempt at this I've seen https://eyurtsev.github.io/kor/index.html https://eyurtsev.github.io/kor/index.html"  
[X Link](https://x.com/hwchase17/status/1634606661137731584)  2023-03-11T17:26Z 88.4K followers, 103.1K engagements


"Integration 2/n: GPT-4 If you are lucky enough to have access to GPT-4 using it in @langchain is quite simple: just update the model name from import ChatOpenAI llm=ChatOpenAI(model_name="gpt-4") http://langchain.chat http://langchain.chat"  
[X Link](https://x.com/hwchase17/status/1635705430860247042)  2023-03-14T18:12Z 88.4K followers, 82.9K engagements


"✨LangChain x Weights & Biases✨ @wandb is the canonical MLOps/experimentation platform. Excited to announce that you can now use it to track your @langchain runs Thanks to Anish Shah for working to get this in Links 👇"  
[X Link](https://x.com/hwchase17/status/1637855557423083520)  2023-03-20T16:36Z 88.4K followers, 70K engagements


"OpenAI recently open-sourced chatgpt-retrieval-plugin a module for creating a retriever based on your own data Can you combine this with @langchain's 40+ DocumentLoaders Yes. Yes you can. Gist for doing so: https://gist.github.com/hwchase17/1429e54879f0249f0a258382d8bd744c https://github.com/openai/chatgpt-retrieval-plugin https://gist.github.com/hwchase17/1429e54879f0249f0a258382d8bd744c https://github.com/openai/chatgpt-retrieval-plugin"  
[X Link](https://x.com/hwchase17/status/1639013354391113728)  2023-03-23T21:16Z 88.4K followers, 177.5K engagements


"🦙LangChain x Llama🦙 Llama integration is finally here Not only with the original llama-cpp library but also with @nomic_ai's gpt4all model (an assistant-style large language model with 800k GPT-3.5-Turbo Generations based on LLaMa) Rather large 🦜🔗0.0.131 release GPT4all model (@nomic_ai) 🦙Llama-cpp model Support for @qdrant_engine local db 🌲Zilliz cloud (@milvusio) Vectorstore support 📧New OutlookMessage Document Loader 🕸New Selenium Document Loader 🪟 Support for SQL views in SQLChain 🧵 Rather large 🦜🔗0.0.131 release GPT4all model (@nomic_ai) 🦙Llama-cpp model Support for"  
[X Link](https://x.com/hwchase17/status/1643264428702601221)  2023-04-04T14:49Z 88.4K followers, 76.9K engagements


"It's been one of the more requested topics for a webinar and it's finally here. 🤖Agents🤖 Featuring: @yoheinakajima (recently doing a lot of work on baby-AGI) @mbusigin (code agents cognosis) Next Wednesday 9am PST: https://www.crowdcast.io/c/46erbpbz609r https://www.crowdcast.io/c/46erbpbz609r"  
[X Link](https://x.com/hwchase17/status/1643694241695006721)  2023-04-05T19:17Z 88.4K followers, 77.6K engagements


"Baby-AGI by @yoheinakajima is taking the world by storm Here's an implementation within the @langchain framework allowing you to easily substitute in other vectorstores and other LLMs Docs: https://python.langchain.com/en/latest/use_cases/agents/baby_agi.html https://python.langchain.com/en/latest/use_cases/agents/baby_agi.html"  
[X Link](https://x.com/hwchase17/status/1645442763536891904)  2023-04-10T15:05Z 88.4K followers, 1M engagements


"🐪CAMEL🐪 Communicative Agents for Mind Exploration of LLM Society This paper shows how to put [--] agents in a sandbox with each other and watch them interact. Now implemented in LangChain (s/o @guohao_li) Original Paper: Docs: https://python.langchain.com/en/latest/use_cases/agents/camel_role_playing.html https://arxiv.org/abs/2303.17760 https://python.langchain.com/en/latest/use_cases/agents/camel_role_playing.html https://arxiv.org/abs/2303.17760"  
[X Link](https://x.com/hwchase17/status/1645834030519296000)  2023-04-11T16:59Z 88.4K followers, 239.9K engagements


"Our webinar on agents starts in [--] hour It's the most popular webinar we've hosted yet so we had to bring in the best possible moderator: @charles_irl Come join Charles myself @ShunyuYao12 @mbusigin and @yoheinakajima for some lively discussion :) https://www.crowdcast.io/c/46erbpbz609r https://www.crowdcast.io/c/46erbpbz609r"  
[X Link](https://x.com/hwchase17/status/1646166299083436032)  2023-04-12T15:00Z 88.4K followers, 108.8K engagements


"How to best build an agent that has access to ALL the ChatGPT Plugins IMO by combining two techniques we recently introduced in @langchain 🔧 Tool Retrieval 🗣 Natural Language APIs Explanation and example notebook in 🧵"  
[X Link](https://x.com/hwchase17/status/1646255350386130945)  2023-04-12T20:53Z 88.4K followers, 131.3K engagements


"Yesterday @karpathy tweeted about using SVMs instead of KNN for retrieval (pros: better results flexibility; cons: takes longer) Today @RLanceMartin implemented it in @langchain 🚀🚀 Play around with it here https://python.langchain.com/en/latest/modules/indexes/retrievers/examples/svm_retriever.html https://python.langchain.com/en/latest/modules/indexes/retrievers/examples/svm_retriever.html"  
[X Link](https://x.com/hwchase17/status/1647328542529843200)  2023-04-15T19:58Z 88.4K followers, 199.6K engagements


"🤖Autonomous Agents & Agent Simulations🤖 Four agent-related projects (AutoGPT BabyAGI CAMEL and Generative Agents) have exploded recently We wrote a blog on they differ from previous @langchain agents and how we've incorporated some key ideas https://blog.langchain.dev/agents-round/ https://blog.langchain.dev/agents-round/"  
[X Link](https://x.com/hwchase17/status/1648703751086022657)  2023-04-19T15:03Z 88.4K followers, 239.9K engagements


"Retrieval for QA systems is hard Vector search is good for capturing semantically similar texts but often queries specify desired attributes like time authorship or other "metadata" fields which vector search is not great at Enter. SelfQueryRetriever"  
[X Link](https://x.com/hwchase17/status/1651617956881924096)  2023-04-27T16:03Z 88.4K followers, 247K engagements


"🔧Structured Tools🔧 Agents are all about the tools you give it Tools in @langchain used to just take a single string input. In our new release tools can now take multiple inputs. We also introduce a new agent type for these tools Blog: 🧵 https://blog.langchain.dev/structured-tools/ https://blog.langchain.dev/structured-tools/"  
[X Link](https://x.com/hwchase17/status/1653781790308597762)  2023-05-03T15:21Z 88.4K followers, 100.7K engagements


"🔀Router Chains🔀 A simple (yet much requested) abstraction that started with a @ShreyaR pr months ago and is finally in @langchain - Router Chain does classification to choose sub chain to use - Call the selected chain with that input Lots of potential use cases 🧵"  
[X Link](https://x.com/hwchase17/status/1654196635113783296)  2023-05-04T18:49Z 88.4K followers, 82.6K engagements


"We just put all the @langchain webinars on Youtube The best part is this now means we can use LangChain (in [--] lines of code) to do question/answering over the LangChain webinars 🤯 How circular Gist: https://gist.github.com/hwchase17/aa78ad6fcb43d84665bdf2cf03afbc8e https://www.youtube.com/channel/UCC-lyoTfSrcJzA1ab3APAgw https://gist.github.com/hwchase17/aa78ad6fcb43d84665bdf2cf03afbc8e https://www.youtube.com/channel/UCC-lyoTfSrcJzA1ab3APAgw"  
[X Link](https://x.com/hwchase17/status/1654964453958651904)  2023-05-06T21:40Z 88.4K followers, 62.9K engagements


"🗺 Plan-and-Execute Agents 🗺 Inspired by BabyAGI and the recent Plan-and-Solve paper we're introducing a new type of @langchain agent We think these are better for more complex tasks at the cost of more calls to the LLM Blog: 🧵 https://blog.langchain.dev/plan-and-execute-agents/ https://blog.langchain.dev/plan-and-execute-agents/"  
[X Link](https://x.com/hwchase17/status/1656327621335195648)  2023-05-10T15:57Z 88.4K followers, 80.2K engagements


"🌟privateGPT🌟 - this is sick I've always had people asking me if it was feasible to use @langchain with open source models and my answer was "at the moment not really." but @ivanmartit did it this is a huge step forward 👏👏👏 https://github.com/imartinez/privateGPT https://github.com/imartinez/privateGPT"  
[X Link](https://x.com/hwchase17/status/1657427349623934976)  2023-05-13T16:47Z 88.4K followers, 489.1K engagements


"How to speed up "chat-your-data" applications while retaining final answer accuracy 🫙Use a cheaper/faster model (gpt-3.5) to create the condensed question 💬Use a better but more expensive model (gpt-4) for final response Thanks to @cristobal_dev for highlighting 🧵"  
[X Link](https://x.com/hwchase17/status/1663576081918742530)  2023-05-30T16:00Z 88.4K followers, 117.4K engagements


"When connecting LLMs to your data semantic search is a great first pass - but to get even better performance you can turn to more advanced techniques: ✒Contextual Compression 🤳Self Query 🕐Time weighting Going live in 30m to talk about this and more https://zilliz.com/event/memory-for-llm-applications-webinarutm_source=partner&utm_medium=referral&utm_campaign=2023-06-08_webinar_retrieval-langchain_zilliz&utm_term=speaker https://zilliz.com/event/memory-for-llm-applications-webinarutm_source=partner&utm_medium=referral&utm_campaign=2023-06-08_webinar_retrieval-langchain_zilliz&utm_term=speaker"  
[X Link](https://x.com/hwchase17/status/1666829939918745600)  2023-06-08T15:30Z 88.4K followers, 69.6K engagements


"The new @OpenAI functions are good for other things besides agents Another killer use case is extracting structured information from unstructured docs We've adding support for extraction AND tagging in @langchain - thanks to @fpingham for code and @jxnlco for review 🧵"  
[X Link](https://x.com/hwchase17/status/1669735823103897600)  2023-06-16T15:56Z 88.4K followers, 319.2K engagements


"Many startups focused on serving open source models have exposed APIs that are interoperable with OpenAI This makes it easy for anyone to try them out by just changing the API base We added better support for this in @langchain in the most recent release (h/t Yuze Ma) 🧵"  
[X Link](https://x.com/hwchase17/status/1673393425456963611)  2023-06-26T18:10Z 88.4K followers, 78K engagements


"📄Documents x LLMs📄 Combining documents with LLMs is a key part of retrieval and chaining We've improved our @langchain reference documentation across the [--] major CombineDocumentsChains and helper functions to help with clarity and understanding of how these work 🧵"  
[X Link](https://x.com/hwchase17/status/1676593263472517121)  2023-07-05T14:05Z 88.4K followers, 135.7K engagements


"🎉New @DeepLearningAI_ class🎉 I had so much fun teaching the last one with @AndrewYNg I had to return for a follow up This one is a deep dive on the most popular applications of LLMs to date: using them to chat with your data What do we cover 👇 https://www.deeplearning.ai/short-courses/langchain-chat-with-your-data/ https://www.deeplearning.ai/short-courses/langchain-chat-with-your-data/"  
[X Link](https://x.com/hwchase17/status/1676607443374075905)  2023-07-05T15:02Z 88.4K followers, 123.8K engagements


"🌳Tree-of-Thought A new reasoning method originally by @ShunyuYao12 implemented in 🦜🧪langchain_experimental by Vadim Gubergrits Paper: Docs: LangSmith Trace: https://smith.langchain.com/public/cf896fa9-0ec8-42e7-83a0-d1f9107b3aaf/r https://python.langchain.com/docs/modules/chains/additional/tot https://arxiv.org/abs/2305.10601 https://smith.langchain.com/public/cf896fa9-0ec8-42e7-83a0-d1f9107b3aaf/r https://python.langchain.com/docs/modules/chains/additional/tot https://arxiv.org/abs/2305.10601"  
[X Link](https://x.com/hwchase17/status/1684595491890487298)  2023-07-27T16:03Z 88.4K followers, 47.2K engagements


"My TED talk is out I talk about what it takes to build "context-aware reasoning applications" including: [--] different ways of providing context to LLMs [--] different types of cognitive architectures https://www.ted.com/talks/harrison_chase_the_magical_ai_assistants_of_the_future_and_the_engineering_behind_them https://www.ted.com/talks/harrison_chase_the_magical_ai_assistants_of_the_future_and_the_engineering_behind_them"  
[X Link](https://x.com/hwchase17/status/1736066556877324692)  2023-12-16T16:51Z 88.4K followers, 82.9K engagements


"❓What is an agent I get asked this question a lot so I wrote a little blog on this topic and other things: - What is an agent - What does it mean to be agentic - Why is agentic a helpful concept - Agentic is new Check it out here: https://blog.langchain.dev/what-is-an-agent/ https://blog.langchain.dev/what-is-an-agent/"  
[X Link](https://x.com/hwchase17/status/1807067248403583461)  2024-06-29T15:03Z 88.4K followers, 58.4K engagements


"yeah deep research is great. but have you ever wanted it open source with swappable models and able to research over your own data GPT-Researcher is exactly that - the leading OSS AI Researcher project https://github.com/assafelovic/gpt-researcher https://github.com/assafelovic/gpt-researcher"  
[X Link](https://x.com/hwchase17/status/1886624151030653120)  2025-02-04T03:53Z 88.4K followers, 71.9K engagements


"coolest client that integrates with MCP that ive seen @hwchase17 You should try @tryhighlight - you can just @ any MCP server from anywhere. Runs natively. With context grounding so it can use what youre looking at as inputs and outputs. https://t.co/ptnSlVX3na @hwchase17 You should try @tryhighlight - you can just @ any MCP server from anywhere. Runs natively. With context grounding so it can use what youre looking at as inputs and outputs. https://t.co/ptnSlVX3na"  
[X Link](https://x.com/hwchase17/status/1897757113885376808)  2025-03-06T21:12Z 88.4K followers, 65K engagements


"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 building agents - What is LangGraph Flavors of agentic frameworks - Agents vs workflows - Declarative vs non-declarative - Agent abstractions - Multi agent Common Questions - What is the value of a framework - As the models get better will everything become agents instead of workflows - What did OpenAI get wrong in their take -"  
[X Link](https://x.com/hwchase17/status/1914016302261506421)  2025-04-20T18:00Z 88.4K followers, 452.1K engagements


"I am not excited about visual workflow builders [--]. Not simple enough for the average user. I believe there should be a simpler way to create modify and adapt no-code agents [--]. Not scalable for complex use cases Wrote a little blog: https://blog.langchain.com/not-another-workflow-builder/ https://blog.langchain.com/not-another-workflow-builder/"  
[X Link](https://x.com/hwchase17/status/1975603633791377920)  2025-10-07T16:46Z 88.4K followers, 98.7K engagements


"🥳Announcing LangChain and LangGraph [---] LangChain and LangGraph [---] versions are now LIVE For both Python and TypeScript Some exciting highlights: - NEW DOCS - LangChain Agent: revamped and more flexible with middleware - LangGraph 1.0: we've been really happy with LangGraph and this is our official stamp of approval - Standard content blocks: swap seamlessly between models Read more about it here: We hope you love it http://blog.langchain.com/langchain-langgraph-1dot0/utm_medium=social&utm_source=linkedin&utm_campaign=q4-2025_october-launch-week_aw"  
[X Link](https://x.com/hwchase17/status/1981030005229670438)  2025-10-22T16:08Z 88.4K followers, 113.3K engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing