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

Harrison Chase posts on X about llm, agents, open ai, 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)  [social networks](/list/social-networks)  [stocks](/list/stocks)  [finance](/list/finance)  [celebrities](/list/celebrities)  [events](/list/events)  [vc firms](/list/vc-firms)  [travel destinations](/list/travel-destinations)  [cryptocurrencies](/list/cryptocurrencies)  [musicians](/list/musicians) 

**Social topic influence**
[llm](/topic/llm), [agents](/topic/agents), [open ai](/topic/open-ai), [ai](/topic/ai), [if you](/topic/if-you), [tools](/topic/tools), [build](/topic/build), [youtube](/topic/youtube), [in the](/topic/in-the), [code](/topic/code)

**Top assets mentioned**
[Alphabet Inc Class A (GOOGL)](/topic/$googl) [Uber Technologies, Inc. (UBER)](/topic/uber)
### Top Social Posts
Top posts by engagements in the last [--] hours

"🚨Biggest @LangChainAI drop yet🚨 ⚡Dynamic zero-shot composition of multiple chains Easily plug in different subchains (google REPL wikipedia DBs) just by telling the router LLM (in English) when and how to use them pip install langchain==0.0.8 https://github.com/hwchase17/langchain/ https://github.com/hwchase17/langchain/"  
[X Link](https://x.com/hwchase17/status/1589648221269557248)  2022-11-07T15:57Z 80.2K followers, [---] engagements


"🪞refine iterates over the pieces of text in a sequence taking the output of the previous step and refining it based on the next document I think the first implementation of this I saw was @jerryjliu0 in GPT-Index. Big shout out to that project"  
[X Link](https://x.com/hwchase17/status/1600882633608568832)  2022-12-08T15:58Z 78.1K followers, [--] engagements


"I wanted to write this to try to synthesize some common patterns I've been seeing It's far from perfect and we'll continue to add to it (hopefully with your help) Big thanks to @sjwhitmore @jerryjliu0 @dwarkesh_sp @cameroncohen @johnjnay for feedback and thoughts"  
[X Link](https://x.com/hwchase17/status/1600882635835772934)  2022-12-08T15:58Z 78.1K followers, [--] engagements


"@YirenLu @floydophone @dagster @LangChain @OpenAI happy to follow up more offline about this. have heard a few requests for longer form generation so could be a pretty interesting case study (theres not a ton in LangChain about it currently although all the building blocks are there)"  
[X Link](https://x.com/hwchase17/status/1612902937407418369)  2023-01-10T20:03Z 79.4K followers, [---] engagements


"Gonna beef up the tutorials for how to create your own Chat-GPT over specific documents with @langchain What types of documents/knowledge bases would people want to have examples for Eg Notion Obsidian webpages etc"  
[X Link](https://x.com/anyuser/status/1622419133949411328)  2023-02-06T02:17Z 88.4K followers, 102.7K 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/anyuser/status/1649186070251311104)  2023-04-20T22:59Z 88.4K followers, 76.7K engagements


"Really excited to be speaking at the @pinecone meetup in San Francisco on April 27th - sign up below Going to be talking all things "retrieval" - document ingestion text splitting contextual compression. what points should I make sure to cover https://www.pinecone.io/events/meetup-op-stack/ https://www.pinecone.io/events/meetup-op-stack/"  
[X Link](https://x.com/hwchase17/status/1649787052878360577)  2023-04-22T14:47Z 62K followers, 12.8K 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/anyuser/status/1650520415608123392)  2023-04-24T15:21Z 88.4K followers, 87.5K 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/anyuser/status/1654196635113783296)  2023-05-04T18:49Z 88.4K followers, 82.6K engagements


"To say I'm excited to discuss LLM applications with @tchutch94 at @Redisinc Days next week would be an understatement It's been awesome to work with Tyler on the open source project and I'm looking forward to a fun conversation :) https://redis.com/redisdays/ https://redis.com/redisdays/"  
[X Link](https://x.com/hwchase17/status/1658890121062809601)  2023-05-17T17:40Z 57.9K followers, 14.4K engagements


"🕸Vertex AI Support🕸 Vertex AI enables usage of PaLM models (and others) with enterprise-grade privacy security and governance Really excited to announce integrations into LangChain for: 🦜LLMs 💬Chat models 🔱Embedding models Across Python🐍 AND JavaScript🎃 🧵"  
[X Link](https://x.com/hwchase17/status/1661756592025649152)  2023-06-29T21:26Z 45.8K followers, 97K engagements


"Is there a better way to spend a Friday night than watching a LangChain Retrieval Webinar featuring @charles_irl @lateinteraction @jobergum No. No there is not. Now uploaded to YouTube so you can easily watch as many times as you want"  
[X Link](https://x.com/hwchase17/status/1662207729153503233)  2023-05-26T21:22Z 42.3K followers, [---] engagements


"LangChain is great for connecting LLMs to your data. even more so when it's stored in an AI-friendly DB like @MindsDB Excited to talking at their AI-Web convention this Thursday"  
[X Link](https://x.com/hwchase17/status/1663571215972179973)  2023-05-30T15:41Z 42.5K followers, [--] engagements


"One of my favorite recent projects is Quivr by @_StanGirard - "your second brain" which utilizes LangChain to store and retrieve unstructured information On the next @LangChainAI Webinar we will try a new format - a deep dive on Quivr. Join us"  
[X Link](https://x.com/hwchase17/status/1665374599054491648)  2023-06-04T15:07Z 45.8K followers, [---] engagements


"releases like these are always fun cause you get a full overview of all the langchain components. vectordb llms embeddings document loaders tools databases"  
[X Link](https://x.com/hwchase17/status/1666877891001319448)  2023-06-08T18:40Z 45.8K followers, [--] engagements


"I participated in a fun webinar with @zilliz_universe on Thursday where I talked about retrieval We did a deep dive on edge cases that arise with semantic search and methods in LangChain to mitigate those Here are the slides:"  
[X Link](https://x.com/hwchase17/status/1667569970291838977)  2023-06-10T16:30Z 45.8K followers, [---] engagements


"🚨Emergency OpenAI Functions Release🚨 pip install langchain==0.0.199 ✅ Support for functions in chat model wrapper ✅ Convert @langchain tools to functions Docs: Next up. I think its time for a new type of agent 😁 https://github.com/hwchase17/langchain/blob/master/docs/modules/agents/tools/tools_as_openai_functions.ipynb https://github.com/hwchase17/langchain/blob/master/docs/modules/agents/tools/tools_as_openai_functions.ipynb"  
[X Link](https://x.com/anyuser/status/1668682373767020545)  2023-06-13T18:10Z 88.4K followers, 309K engagements


"Lots of applications use the RAG style framework with @LangChainAI and @pinecone But how can you trouble shoot those applications Enter @arizeai Excited to be doing a LangChain x Pinecone x Arize webinar on all this (feat @RLanceMartin) Sign up 👇"  
[X Link](https://x.com/hwchase17/status/1671566668257198081)  2023-06-21T17:12Z 40.7K followers, 17.3K engagements


"I like this a lot - a higher level wrapper around langchain to make a specific use case (LLM-powered bots over a dataset) easier to create What other use cases deserve higher level wrappers around langchain"  
[X Link](https://x.com/hwchase17/status/1672738339915698176)  2023-06-24T22:47Z 45.8K followers, [---] engagements


"Pretty good spec for langchain chatbot wrapper (langchat) Let use easily customize LLM used system message thats it Whos gonna build"  
[X Link](https://x.com/hwchase17/status/1672818684321218562)  2023-06-25T04:07Z 45.8K followers, [--] engagements


"anyone tried this out with a langchain (qa over docs sql) yet"  
[X Link](https://x.com/hwchase17/status/1674519524870328320)  2023-06-29T20:45Z 45.8K followers, [--] engagements


"🦜🔗 LangChain x @AirbyteHQ x @dagster Load documents from one of @AirbyteHQ's 300+ sources split/embed/store it with @LangChainAI. all orchestrated by @dagster Excited to share this end to end guide"  
[X Link](https://x.com/hwchase17/status/1676670452448612353)  2023-07-05T19:12Z 45.8K followers, [--] engagements


"awesome stuff showing how to build slack bots using langchain and @JinaAI_"  
[X Link](https://x.com/hwchase17/status/1676958915475705858)  2023-07-06T14:18Z 45.8K followers, [--] engagements


"Some big improvements to the integrations page including new categories (callbacks) and new pages (top/trending/new)"  
[X Link](https://x.com/hwchase17/status/1676969918309502979)  2023-07-06T15:02Z 40.5K followers, [----] engagements


"LangChain "Characters" Webinar One of the more creative (and fun) use cases of LLMs is having them impersonate a character Real world use cases can include NPCs for video games Join us this Thusday to hear a fun discussion on how best to do this"  
[X Link](https://x.com/anyuser/status/1678459468437397504)  2023-07-11T00:02Z [--] followers, 18.9K engagements


"Ever since the first LangChain meetup we've had a high level of interest in demos In an attempt to better showcase what the community is building we're dedicating time and space at our next meetup for a lot of demos Sign up here:"  
[X Link](https://x.com/hwchase17/status/1678808635777126402)  2023-07-11T16:49Z 43.6K followers, [--] engagements


"There are different use cases for AI characters and creative use cases like this are a really fun one Excited to have @AkashSamant4 join our @LangChainAI Characters Webinar and hear more about it Sign up for it below (happening this Thursday)"  
[X Link](https://x.com/hwchase17/status/1678826338466275329)  2023-07-11T18:01Z 40.2K followers, [----] engagements


"From the beginning @streamlit has been a fantastic tool for LangChain developers One of the first examples we released used Streamlit as the UI Excited to share some of the stuff we've been working on and thinking about"  
[X Link](https://x.com/anyuser/status/1678869121168216065)  2023-07-11T20:49Z [--] followers, [--] engagements


"@AceBryanLiu @0xSamHogan @LangChainAI Build an agent with neo4j"  
[X Link](https://x.com/hwchase17/status/1679306276701106177)  2023-07-13T02:27Z 45.8K followers, [---] engagements


"There's huge demand for AI companions as shown by @stuffyokodraws AI companion kit We recently hosted a webinar (w/ @Metropolize_AI @AkashSamant4 @mmmbchang) on creating AI characters with LangChain - now on YouTube for some fun Friday viewing"  
[X Link](https://x.com/hwchase17/status/1679959211365773312)  2023-07-14T21:01Z 45.8K followers, [---] engagements


"🦜🔗Code Interpreter API ChatGPT's code interpreter is the hottest thing in the streets. A new project by @Shroominic takes a stab at recreating that functionality locally using OpenAI's apis Uses CodeBox - a sandboxed env with simple fileIO functionality Blog & GitHub 👇"  
[X Link](https://x.com/anyuser/status/1680620252433117184)  2023-07-16T16:47Z 88.4K followers, 152.5K engagements


"clever hiding of langchain behind a pydantic model"  
[X Link](https://x.com/hwchase17/status/1680620853728518144)  2023-07-24T01:07Z 41.9K followers, [----] engagements


"🦜🔗LangChain Gradio Custom QA Over Docs New repo showing how to use the new @Gradio chatbot release to create an application to chat with your docs Crucially does NOT use ConversationalRetrievalQA chain but rather only individual components to show how to customize 🧵"  
[X Link](https://x.com/hwchase17/status/1681096742652215296)  2023-07-24T01:07Z 42.3K followers, 30.1K engagements


"Uses @trychroma for vector DB @cohere for embeddings @OpenAI for text generation (though with LangChain you can easily swap those out) Thanks to @abidlabs for help Repo link:"  
[X Link](https://x.com/hwchase17/status/1681096743688237061)  2023-07-18T00:21Z 40.5K followers, [--] engagements


"Quickly highlighting some awesome end-to-end examples (complete with UI) leveraging @LangChainAI and @streamlit: basic_streaming mrkl_demo minimal_agent search_and_chat chat_with_documents What examples should we add next Repo:"  
[X Link](https://x.com/hwchase17/status/1681098960989978626)  2023-07-24T01:07Z 45.8K followers, 40K engagements


"A year ago there wasn't enough "tinkering" happening (h/t @natfriedman) 🦜🔗That's why we started LangChain Now there aren't enough applications making their way from prototype to production 🦜🛠That's why we're launching LangSmith"  
[X Link](https://x.com/hwchase17/status/1681349680373075969)  2023-07-18T17:06Z 46K followers, [---] engagements


"One of the cool things about LangSmith is it integrates seamlessly with all projects built on top of or integrated with LangChain Particularly useful for seeing what calls are actually getting made + the prompts that are used Here is the call that PandasAI makes:"  
[X Link](https://x.com/hwchase17/status/1681424858507272192)  2023-07-18T22:05Z 45.8K followers, [--] engagements


"🔎Code Interpreter Insights @Shroominic released an incredible open source implementation of code interpreter this weekend: Since its built on top of LangChain you can easily visualized everything that's going on under the hood"  
[X Link](https://x.com/hwchase17/status/1681456818692784131)  2023-07-19T00:12Z 42.3K followers, [---] engagements


"🧬How will LLMs and GenAI will transform biotech I don't know but I'm excited to learn at our next more focused meetup: AI x BIO Lightning talks with @bitsinbio on July 27th We're looking for more speakers at this intersection so reach out"  
[X Link](https://x.com/hwchase17/status/1681716026780655616)  2023-07-19T17:22Z 43K followers, [--] engagements


"Excited to support this in the newest langchain release"  
[X Link](https://x.com/hwchase17/status/1681777324373708800)  2023-07-19T21:25Z 45.8K followers, [--] engagements


"👏👏An AWESOME trace by @kreneskyp generating an entire app with a single SequentialChain We add a new #cool-langsmith-traces channel in our discord to share fun things like this. How crazy can it get https://smith.langchain.com/public/92835ead-e1da-45b2-b99d-4ff1a0f6c301/r https://smith.langchain.com/public/92835ead-e1da-45b2-b99d-4ff1a0f6c301/r"  
[X Link](https://x.com/hwchase17/status/1682042700948987909)  2023-07-20T15:00Z 61.8K followers, 15.6K engagements


"langchain ecosystem continues to grow. the first tweet can't even contain all the new things Check out the full list of new ones here: My favorite The browserless web loader:"  
[X Link](https://x.com/hwchase17/status/1682060122112274432)  2023-07-24T01:07Z 45.8K followers, 13K engagements


"This is awesome AI programmer consisting of a team of autonomous agents Here's a trace showing what's going on under the hood: Really interesting to see how the repo structure is injected in the prompt"  
[X Link](https://x.com/hwchase17/status/1682081258694918144)  2023-07-20T19:23Z 45.8K followers, 36.9K engagements


"This is great. Really ties together the whole development/debugging life cycle Would be cool to get these for some UIs in as well"  
[X Link](https://x.com/hwchase17/status/1682173695975100417)  2023-07-24T01:07Z 43K followers, [----] engagements


"This is the next midjourney Been playing with this for a week now and its addicting Fully AI generated music 🤯"  
[X Link](https://x.com/hwchase17/status/1682217958163234817)  2023-07-21T17:20Z 45.8K followers, 50.4K engagements


"Some AWESOME community contributions to the @LangChainAI x @streamlit repo 🧱Chat-SQL App by @meshashank93: 🐼Pandas Agent by @maraza24: See those and more here:"  
[X Link](https://x.com/hwchase17/status/1682416140104900608)  2023-07-21T17:23Z 40.8K followers, 12.7K engagements


"🧪langchain_experimental In an effort to make langchain leaner more focused and safer we are moving select chains to a separate package on 7/28 Big thanks to folks like @BoazWasserman @OrRaz6 Justin Flick for pushing on the safety part There will be some breaking changes 🧵"  
[X Link](https://x.com/hwchase17/status/1682785686729166848)  2023-07-22T16:12Z 43.2K followers, 41.9K engagements


"In particular users of the SQLDatabaseChain will need to migrate to langchain_experimental Read more about the motivation and track the progress here: Read about how to migrate here: Comments feedback and questions welcome"  
[X Link](https://x.com/hwchase17/status/1682785687895171072)  2023-07-22T16:12Z 45.8K followers, [----] 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/anyuser/status/1683153581200019456)  2023-07-23T16:34Z 88.4K followers, 124.3K engagements


"Something I was working on at the @agihouse_org hackathon over the weekend The LLMChain (prompt template + llm + output parser) is a core modular component and is where most customization happens. Understanding this is crucial for understanding LangChain"  
[X Link](https://x.com/hwchase17/status/1683507598295961603)  2023-07-24T16:01Z 45.8K followers, 16.7K engagements


"The whole @RealCharAI team is awesome Really excited to have this opportunity to help highlight their work Play around with their GitHub here: Read blog here: https://blog.langchain.dev/realchar-x-langsmith-ai-companions/ https://github.com/Shaunwei/RealChar we're super excited about the future of AI characters & companions & @RealCharAI may be the most complete and exciting OSS AI character framework out there they created something really sophisticated & fun to use all with open source tools Full story: https://t.co/2265GNFbGB https://blog.langchain.dev/realchar-x-langsmith-ai-companions/"  
[X Link](https://x.com/hwchase17/status/1683536002441428993)  2023-07-24T17:53Z 65.7K followers, [----] engagements


"🤗HuggingGPT Another installment to langchain_experimental🧪 Thanks to @itricktreat you can now use an agent which connects to a variety of @huggingface models in @LangChainAI Based on the paper here: Docs:"  
[X Link](https://x.com/hwchase17/status/1683537213114699777)  2023-07-24T17:58Z 45.8K followers, 32.6K engagements


"A great read on summarizing webpages from our friends at @UnstructuredIO"  
[X Link](https://x.com/hwchase17/status/1683883055927242758)  2023-07-30T01:42Z 45.8K followers, [----] engagements


"🦙Llama [--] support One really cool thing about this project - we made support for Llama [--] a first class citizen This includes prompts specifically tuned for the Llama [--] model We'll be talking about this heavily in our webinar later today 👇"  
[X Link](https://x.com/hwchase17/status/1684242282508320769)  2023-07-30T01:42Z 45.8K followers, 14.1K engagements


"💬Few shot prompt templates for chat models As you look to improve performance few shot example is one of the easiest and best knobs to turn We're working on a bunch of stuff to make this easier s/o @WHinthorn and @veryboldbagel for improving the UX here"  
[X Link](https://x.com/hwchase17/status/1685295110245040128)  2023-07-30T04:36Z 45.8K followers, 19.2K engagements


"🕸XML Agent Some language models (like @AnthropicAI's Claude) are particularly good at reasoning/writing XML We recently added a new type of agent that uses XML when prompting. pip install langchain==0.0.248 Docs: https://python.langchain.com/docs/modules/agents/agent_types/xml_agent https://python.langchain.com/docs/modules/agents/agent_types/xml_agent"  
[X Link](https://x.com/hwchase17/status/1686025584034430977)  2023-07-31T14:46Z 62K followers, 19.9K engagements


"💬LangChain Expression Language This is a big change. It allows for: 🤝Unified streaming async and batch interfaces across all chains ⛓Easy declarative composition of chains making them easier to understand and modify 🧵of resources below"  
[X Link](https://x.com/anyuser/status/1686425322261213184)  2023-08-01T17:15Z [--] followers, 35.7K engagements


"A 🧵on examples of using our new LangChain Expression Language to rewrite some of our most popular chains Benefits: it's very clear what's going on under the hood and (most importantly) how to modify them 👇"  
[X Link](https://x.com/hwchase17/status/1686511196802007040)  2023-08-01T22:56Z 45.8K followers, 29.3K engagements


"Router You can route between various downstream chains depending on the output of a previous one"  
[X Link](https://x.com/hwchase17/status/1686511211280728068)  2023-08-01T23:19Z 45.8K followers, [---] engagements


"This was a really cool project Very impressive and novel way to combine LLMs with external data"  
[X Link](https://x.com/hwchase17/status/1686600624711020544)  2023-08-02T19:06Z 43.6K followers, 15.3K engagements


"A good overview of some samples of the new LangChain syntax As a reminder we're going live in [--] minutes with the one and only @nfcampos to discuss the motivation the interface and some examples https://www.crowdcast.io/c/ckw1tydg29er Rewritten several @LangChainAI Expression Language samples with @ValDotTown for quick references check them out in the following thread 🔗 https://www.crowdcast.io/c/ckw1tydg29er Rewritten several @LangChainAI Expression Language samples with @ValDotTown for quick references check them out in the following thread 🔗"  
[X Link](https://x.com/hwchase17/status/1686763056326684672)  2023-08-02T15:37Z 61.2K followers, [----] engagements


"Chat-Your-Data Update Our starter repo for creating a chatbot over your data has got an update Big thanks to @bllchmbrs for updating it (new chains new LLMs) The space moves fast is great to have community members like Bill keeping us up-to-date"  
[X Link](https://x.com/hwchase17/status/1687485291932872704)  2023-08-05T06:15Z 40.3K followers, 30.2K engagements


"Had a fun discussion with @nfcampos on Wednesday about our new LangChain Expression Language We also took some awesome questions from the audience Check it out on YouTube now for some fun Friday viewing:"  
[X Link](https://x.com/hwchase17/status/1687489066873376768)  2023-08-04T15:42Z 42.5K followers, [----] engagements


"🌲Advanced Retrieval Webinar @LangChainAI x @trychroma x @UnstructuredIO Retrieval is a key part of most GenAI systems. And there is a lot of nuance to it Excited to bring together some leaders from across the stack to discuss it live next week 👇"  
[X Link](https://x.com/hwchase17/status/1687536626380210176)  2023-08-04T18:50Z 43.2K followers, 19.8K engagements


"Joris's stuff is always awesome We've been playing around with teaching bots internally curious to compare to this"  
[X Link](https://x.com/hwchase17/status/1687539813803827205)  2023-08-04T21:44Z 45.8K followers, [----] engagements


"💪Deep dive on retrieval Excited to have the opportunity to talk about the advanced retrieval tools we're building: - smart text splitting - self-query - retrieval agents Even more excited to be joined by @UnstructuredIO @trychroma. should be fun"  
[X Link](https://x.com/hwchase17/status/1687854977891315713)  2023-08-05T15:56Z 45.8K followers, 16.5K engagements


"🧬CSV Question Answering Most QA applications deal with asking questions of unstructured data. But a lot of data is in tabular (excel csv) format This has always been tricky to get right. We want to improve that But we need your help 👇"  
[X Link](https://x.com/hwchase17/status/1687874486677430272)  2023-08-05T17:13Z 45.8K followers, 129.9K engagements


"@eerac Def thats the goal Its all using titanic for now so no privacy concerns there"  
[X Link](https://x.com/hwchase17/status/1688020081127968768)  2023-08-06T04:47Z 45.8K followers, [---] engagements


"@anuragm75160136 Weve got a lot of graph qa chains is that what you meant"  
[X Link](https://x.com/hwchase17/status/1688025334862766080)  2023-08-06T03:12Z 45.8K followers, [--] engagements


"🏁Self-Query The idea of splitting a user query into a semantic search term as well as a metadata filter One of the more clever retrieval methods in @LangChainAI courtesy of @veryboldbagel Excited to talk about it in our webinar next week"  
[X Link](https://x.com/hwchase17/status/1688203776589434880)  2023-08-06T15:02Z 45.8K followers, 32.2K engagements


"Gonna check this out this afternoon Been spending a lot of time creating langchain x @streamlit apps - if this can speed that up would be great"  
[X Link](https://x.com/hwchase17/status/1688270272686919680)  2023-08-06T19:26Z 45.8K followers, 25.7K engagements


"It's really hard to evaluate LLM applications The most direct way is to do so is to gather feedback from the end user Here's an in depth walkthrough of using LangSmith to do that. Feedback is associated with traces so you can easily debug bad results"  
[X Link](https://x.com/hwchase17/status/1688980646113447939)  2023-08-08T18:29Z 45.8K followers, 27.3K engagements


"Bringing robust scheduling and orchestration to unstructured data ingestion with @AirbyteHQ Transformations powered by LangChain"  
[X Link](https://x.com/hwchase17/status/1689015258474745857)  2023-08-08T20:46Z 45.8K followers, 11.3K 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/anyuser/status/1689015952623771648)  2023-08-08T20:49Z 88.4K followers, 144.3K engagements


"This is a great example code snippet for creating a QA bot that streams via FastAPI The LangChain Expression Language makes this much simpler than before (more accurately the standard streaming interface that all LCEL object have makes this simpler)"  
[X Link](https://x.com/hwchase17/status/1689048400099254273)  2023-08-08T22:58Z 45.8K followers, [----] engagements


"@CamAHutchison @ChrisBoraTweets I ended up going a slightly different direction (in terms of implementation): ParentDocumentRetriever - retrieves the parent document for a given chunk. not exactly the same but close"  
[X Link](https://x.com/hwchase17/status/1689068420510490624)  2023-08-09T00:17Z 40.3K followers, [---] engagements


"👪Parent Document Retriever A new retrieval algorithm that: ✂Creates small chunks (to allow embeddings to have semantic meaning) 🧱Fetches the PARENT documents those chunks came from (to capture full context) Parent documents can be either raw documents or larger chunks 👇"  
[X Link](https://x.com/hwchase17/status/1689301894769094656)  2023-08-09T15:45Z 43.2K followers, 42.9K engagements


"There was always a balance between small enough chunks to allow the embeddings to capture the semantic meaning of a chunk while at the same time wanting them to be long enough to have the full context This helps strike this balance Docs:"  
[X Link](https://x.com/hwchase17/status/1689301896211898368)  2023-08-09T15:45Z 45.8K followers, [----] engagements


"Can also be thought of as similar to fetching the chunk before/after the most semantically similar chunk This takes advantage of the fact that context windows are getting longer and longer This would have been less plausible a few months ago"  
[X Link](https://x.com/hwchase17/status/1689301897335984128)  2023-08-09T15:45Z 61.2K followers, [----] engagements


"We'd been noodling on this concept for a bit but spurred into action by @ChrisBoraTweets yesterday: What other retrieval algorithms would be good to add Docs: https://python.langchain.com/docs/modules/data_connection/retrievers/parent_document_retriever https://x.com/ChrisBoraTweets/status/1689018380584439808s=20 @hwchase17 One feature I wish LangChain had is the ability to insert neighboring chunks of the top K matching chunks from the retriever into the context. If LangChain included this feature using 3-sentence chunks would work exceptionally well for RAG use cases"  
[X Link](https://x.com/hwchase17/status/1689301898342608896)  2023-08-09T15:45Z 62K followers, [----] engagements


"Lots of fun stuff in LangSmith - debugging testing evaluation monitoring And you can use it all even if you arent using LangChain Great new guide"  
[X Link](https://x.com/hwchase17/status/1689375361904291840)  2023-08-09T20:37Z 45.8K followers, 18.1K engagements


"New vercel template and cookbook for agents - show the intermediate steps in a clear way"  
[X Link](https://x.com/hwchase17/status/1689663373267566592)  2023-08-10T15:41Z 45.8K followers, [----] engagements


"🎥"Production Ingestion" Webinar In order to have good retrieval you need to ingest data properly and at scale. Excited to announce that will be the focus our next webinar We'll be joined by @AirbyteHQ and @sweep__ai. Register below:"  
[X Link](https://x.com/hwchase17/status/1689690398430793728)  2023-08-10T17:29Z 45.8K followers, [----] engagements


"Friday viewing: "Advanced Retrieval" Webinar I had a great time discussing: - Importance of preprocessing @mrobinson0623 of @UnstructuredIO - Different retrieval methods to power RAG applications - What lies beyond simple RAG @atroyn of @trychroma"  
[X Link](https://x.com/hwchase17/status/1690021755241521152)  2023-08-11T15:26Z 40.2K followers, 27.9K engagements


"🚀We just sent out the second edition of our newsletter and its fun to see the pace of the community We added two new sections ("Use-cases we love" and "Thank You's") to highlight this We also cover LangSmith updates like team support and our new cookbook repo"  
[X Link](https://x.com/hwchase17/status/1690034770754015233)  2023-08-11T16:17Z 40.3K followers, [----] engagements


"If you're not on it you can read it here: You can also join the 20k+ subscribers by signing up here:"  
[X Link](https://x.com/hwchase17/status/1690034772259766272)  2023-08-11T16:17Z 45.8K followers, [----] engagements


"a great overview of some langchain apps to take inspiration from we've been running @LangChainAI in prod now for a couple months before we launched we took inspo from a few other in-production langchain apps that @thepaulthomson has helpfully compiled into a blog post https://t.co/Nleax27mWM we've been running @LangChainAI in prod now for a couple months before we launched we took inspo from a few other in-production langchain apps that @thepaulthomson has helpfully compiled into a blog post https://t.co/Nleax27mWM"  
[X Link](https://x.com/hwchase17/status/1690379864959066112)  2023-08-12T15:09Z 62K followers, [----] engagements


"This covers: - logging feedback - using langsmith WITHOUT langchain - evaluating a question/answering application end-to-end - evaluating applications wheres theres not static test cases A GREAT saturday reading :)"  
[X Link](https://x.com/hwchase17/status/1690481326313684992)  2023-08-12T21:52Z 45.8K followers, [--] engagements


"Went through some iterations did some testing and I've released a new version of this app Would love any more help testing/providing feedback (thanks to everyone who helped already). Going to do a write up and open-source all learnings next week"  
[X Link](https://x.com/hwchase17/status/1690776862203748352)  2023-08-13T17:26Z 45.8K followers, 32.5K engagements


"webinar link: https://www.crowdcast.io/c/kkn1f3ewlqvq Last week we shared a blog by @AirbyteHQ on creating a slack bot to chat with your GitHub data It was powered by new Airbyte vector database destination Excited to have them on for our next webinar to talk about this and more https://t.co/pFYoiLzEdW https://www.crowdcast.io/c/kkn1f3ewlqvq Last week we shared a blog by @AirbyteHQ on creating a slack bot to chat with your GitHub data It was powered by new Airbyte vector database destination Excited to have them on for our next webinar to talk about this and more https://t.co/pFYoiLzEdW"  
[X Link](https://x.com/hwchase17/status/1690777161748398080)  2023-08-13T17:27Z 61.8K followers, [----] engagements


"🚀Retrieving from multiple sources Oftentimes your application may want to retrieve from multiple sources. When you do this you want to make sure to do that in parallel as much as possible The new LangChain Expression Language supports this out of the box 🧵"  
[X Link](https://x.com/hwchase17/status/1690780072830201856)  2023-08-13T17:39Z 43.6K followers, [---] engagements


"We've added a notebook showing an example of retrieving from multiple sources - in this case one of the them is a SQL database another is a vectorstore"  
[X Link](https://x.com/hwchase17/status/1690780074638032896)  2023-08-13T17:39Z 41K followers, [----] engagements


"If you do these operations sequentially they can start to add up That's why it's important to do them in parallel One nice attribute of the LangChain Expression Language is it tries to do things in parallel where ever possible"  
[X Link](https://x.com/hwchase17/status/1690780077041340416)  2023-08-13T17:39Z 45.8K followers, [--] engagements


"In this case we have a RunnableMap (the dictionary) For these operations they are run in parallel See our guide on that here: https://python.langchain.com/docs/guides/expression_language/interface#parallelism https://python.langchain.com/docs/guides/expression_language/interface#parallelism"  
[X Link](https://x.com/hwchase17/status/1690780078131855360)  2023-08-13T17:39Z 62K followers, [----] engagements


"Really cool to be able to see the logs & complex prompts that went into the Generative Agents simulation"  
[X Link](https://x.com/hwchase17/status/1690830749921083392)  2023-08-13T21:00Z 45.8K followers, [----] engagements


"@douglasqian dm'ed"  
[X Link](https://x.com/hwchase17/status/1690878624835174400)  2023-08-14T00:10Z 45.8K followers, [--] engagements


"Excited to collaborate with @assaf_elovic to bring multiple LLM providers + an integration with LangSmith to GPT Researcher - the best research agent I know of I'd highly encourage folks to try it out - it's very impressive"  
[X Link](https://x.com/hwchase17/status/1691084934113288192)  2023-08-14T13:50Z 45.8K followers, 11.1K engagements


"Excited to be speaking at AWS Gen AI day Lot's of awesome integrations with AWS platform was fun to highlight them all https://genaiday.virtual.awsevents.com/registertrk=57dbe57a-49c3-432b-a93b-0988d5011812&sc_channel=sm https://genaiday.virtual.awsevents.com/registertrk=57dbe57a-49c3-432b-a93b-0988d5011812&sc_channel=sm"  
[X Link](https://x.com/hwchase17/status/1691101620376117248)  2023-08-14T14:57Z 57.1K followers, [----] engagements


"We introduced the ParentDocumentRetriever last week to strike a balance between: - using small chunks during indexing - passing larger chunks to the LLM Great overview and diagram by @clusteredbytes"  
[X Link](https://x.com/hwchase17/status/1691179199594364928)  2023-08-14T20:05Z 45.8K followers, 59.5K engagements


"@usman_akinyemi @LangChainAI this is pretty easily doable - you can give an agent access to multiple retrievers"  
[X Link](https://x.com/hwchase17/status/1691213005814738945)  2023-08-14T22:19Z 43.2K followers, [--] engagements


"🚢Benchmarking Question/Answering Over CSV Data Deep dive on improving an application that does question answering over CSV data: 📜3000 word blog post 🎥30min video 🛌Open sourced eval data 🎬Open sourced code for gathering feedback 🤖Open sourced final agent code 🧵"  
[X Link](https://x.com/hwchase17/status/1691464459456495616)  2023-08-15T14:58Z 45.8K followers, 61.1K engagements


"Blog: YouTube: Code & data used: Now for a quick thread:"  
[X Link](https://x.com/hwchase17/status/1691464461167763457)  2023-08-15T14:58Z 45.8K followers, [----] engagements


"Then we could go about improving it Our final solution was a custom agent with two tools: - Python REPL (for doing analysis) - Retriever (for dealing with text data present) We're excited to see other's solutions for this as well"  
[X Link](https://x.com/hwchase17/status/1691464467190804481)  2023-08-15T14:58Z 43.6K followers, [----] engagements


"Recap of important links: Blog: YouTube: Code & data used: We had a lot of fun doing this and learned a lot - we're going to do it for more tasks Up next: SQL"  
[X Link](https://x.com/hwchase17/status/1691464468684050432)  2023-08-15T14:58Z 42K followers, [----] engagements


""OpenAI said before that evaluating their dataset was one of the major factors for the huge jump from ChatGPT3.5 to [--]. Let's do the same for our own applications" @SimonNom1 expressed better than I could why we're focusing so much on end-to-end evaluations"  
[X Link](https://x.com/hwchase17/status/1691472764161630208)  2023-08-15T15:31Z 45.8K followers, [----] engagements


"If you enjoyed our text splitter playground from last week then you'll want to check out @sweep__ai's code splitter demo And then you'll want to register for our webinar with them (tmrw)"  
[X Link](https://x.com/hwchase17/status/1691476843306598400)  2023-08-15T15:48Z 40.2K followers, [----] engagements


"🧮Benchmarking SQL Question Answering The next end-to-end application we'll be looking to add benchmarking for is questions over a SQL database Similarly to last time we need your help 👇"  
[X Link](https://x.com/hwchase17/status/1691514638083715076)  2023-08-15T18:18Z 40.2K followers, 17.7K engagements


"Access the app here: Please leave feedback on questions that the current implementation does a particularly good or bad job at answering If you want to see the underlying code its at this repo:"  
[X Link](https://x.com/hwchase17/status/1691514639597858818)  2023-08-15T18:18Z 45.8K followers, [----] engagements


"@usman_akinyemi @LangChainAI @LangChain i would probably tell it explicitly in the system message what you just wrote - "if you do not find a person in the prof search in founder" or someting also would use OPENAI_FUNCTIONS agent"  
[X Link](https://x.com/hwchase17/status/1691585980829167699)  2023-08-15T23:01Z 45.8K followers, [--] engagements


"💬ChatLangChain Improvements We're benchmarking a bunch of retrieval and agent methods for our "chat langchain" app Interact with the new beta version here: Expect rapid improvements and everything (data retrieval algorithms prompts) to be OSS"  
[X Link](https://x.com/hwchase17/status/1691876849750782150)  2023-08-16T18:17Z 45.8K followers, 10.8K engagements


"An underrated aspect of LLMs is using them for structured data extraction Extracting knowledge triplets is a great use case I gave it our most recent blog post about @MultiON_AI and it came up with the below - how did it do @DivGarg9 https://blog.langchain.dev/multion-x-langchain-powering-next-gen-web-automation-navigation-with-ai/ Did you ever want to extract knowledge graphs using LLM function calling No Well here's a @streamlit app where you can play around with various inputs. E.g. feed it the Barbie plot gpt-3.5 w/ function calling extracts graph triples. Give it a try:"  
[X Link](https://x.com/hwchase17/status/1691890074563260548)  2023-08-16T19:10Z 61.2K followers, 45.1K engagements


"@ZacharyRoyals1 @MultiON_AI @DivGarg9 ive got some ideas anything specific you have in mind"  
[X Link](https://x.com/hwchase17/status/1691925102542336265)  2023-08-16T21:29Z 45.8K followers, [---] engagements


"@lgertel always was:"  
[X Link](https://x.com/hwchase17/status/1692212125429805143)  2023-08-17T16:29Z 45.8K followers, [--] engagements


"🧮LangChain "RAG Evaluation" Webinar RAGAS is an open-source evaluation framework for your Retrieval Augmented Generation (RAG) pipelines I'm VERY excited to be doing a webinar with them next week RAGAS Repo: Webinar:"  
[X Link](https://x.com/hwchase17/status/1692220493657485674)  2023-08-17T17:03Z 40.3K followers, 23.2K engagements


"🌲RAG Evaluation Webinar Last week @WHinthorn added a fantastic cookbook using LangSmith to test RAG systems: Come listen to him talk about this (and more) during our evaluation webinar next week Register for the webinar here:"  
[X Link](https://x.com/hwchase17/status/1692945772566945996)  2023-08-19T17:05Z 42.5K followers, 11.7K engagements


"Before you can retrieve data you first need to ingest it Chatted in depth about this during our most recent webinar with @AirbyteHQ and @sweep__ai - check it out on YouTube here:"  
[X Link](https://x.com/hwchase17/status/1693290795858907469)  2023-08-20T15:56Z 45.8K followers, 10.7K engagements


"BG Embeddings are at the top of several emebedding leaderboards - awesome guide by @umitmertcakmak showing how to use them in langchain"  
[X Link](https://x.com/hwchase17/status/1693311353484984806)  2023-08-20T17:17Z 43.2K followers, [----] engagements


"📄Amazon Textract Integration Loading PDFs has been one of the biggest areas of questions for folks using LangChain to connect to their data That's why I'm excited to highlight our recently-added integration with Amazon Textract https://python.langchain.com/docs/integrations/document_loaders/pdf-amazonTextractPDFLoader https://python.langchain.com/docs/integrations/document_loaders/pdf-amazonTextractPDFLoader"  
[X Link](https://x.com/hwchase17/status/1693672309784182953)  2023-08-21T17:12Z 61.9K followers, [----] engagements


""This maaaay be a hackathon or a series of workshops or just an extended hangout session schedule TBD" Either way this is going to be a lot of fun Looking forward to learning from some awesome speakers"  
[X Link](https://x.com/hwchase17/status/1693782173533499655)  2023-08-22T00:28Z 40.7K followers, [----] engagements


"Airbyte has 300+ document loaders - this will make it easier to use them in pure python - no need to set up the Airbyte server Bring the necessary context into your LangChains"  
[X Link](https://x.com/hwchase17/status/1694011548753408416)  2023-08-22T15:40Z 45.8K followers, 24.5K engagements


"If you want to explore finetuning LLaMA2 we'll be talking at and helping out with a [--] day in-person event focused explicitly on finetuning OSS models Hopefully this guide will come in handy RSVP here (s/o @swyx and @NaderLikeLadder for organizing):"  
[X Link](https://x.com/hwchase17/status/1694403173623218229)  2023-08-23T17:36Z 45.8K followers, 11.7K engagements


"It's always insightful to hear from actual application developers on how they approach evaluation That's why I'm excited to have @pedro_pxp on our webinar tomorrow that is focused on evaluating RAG systems"  
[X Link](https://x.com/hwchase17/status/1694404803210977769)  2023-08-23T17:42Z 45.8K followers, [----] engagements


"💪RAGAS is an awesome open-source framework for evaluating RAG systems 🤝We recently integrated LangSmith RAGAS for optimal ease of evaluation 🎥We're also doing a webinar on RAG evaluation with them in [--] minutes Link to webinar and blog 👇"  
[X Link](https://x.com/hwchase17/status/1694740124687548714)  2023-08-24T15:55Z 45.8K followers, 22.8K engagements


"Blog: Webinar: Hope to see you there https://www.crowdcast.io/c/bnx91nz59cqq https://blog.langchain.dev/evaluating-rag-pipelines-with-ragas-langsmith/ https://www.crowdcast.io/c/bnx91nz59cqq https://blog.langchain.dev/evaluating-rag-pipelines-with-ragas-langsmith/"  
[X Link](https://x.com/hwchase17/status/1694740125622894760)  2023-08-24T15:55Z 61.9K 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/anyuser/status/1695078249788027071)  2023-08-25T14:18Z 88.4K followers, 134.3K engagements


"The similarity between them all lies in the retrieval step During retrieval the query is compared against the many vectors. Each of these vectors importantly links to a single "parent" document. After retrieval those parent documents are deduplicated and returned"  
[X Link](https://x.com/hwchase17/status/1695078253973938561)  2023-08-25T14:18Z 45.8K followers, [---] engagements


"Importantly with all of these you don't have to choose just one You can use multiple or use multiple and still embed the original document See the full documentation here:"  
[X Link](https://x.com/hwchase17/status/1695078255173517666)  2023-08-25T14:18Z 45.8K followers, [----] engagements


"❓Hypothetical questions This flips the table a bit: generate a list of hypothetical questions that should lead to this document being retrieved Benefits: [--]. Splits the concepts up into unique parts [--]. Puts embeddings in the question/query space (reverse HyDE)"  
[X Link](https://x.com/hwchase17/status/1695078258663129308)  2023-08-25T14:18Z 45.8K followers, [---] engagements


"🖐Manually specified text snippets One of the more underrated parts of this I think is that you specify text snippets/embeddings that SHOULD lead to this document being retrieved A user question misses retrieving this No problem Add it and next time it won't"  
[X Link](https://x.com/hwchase17/status/1695078259682423108)  2023-08-25T14:18Z 45.8K followers, [----] engagements


"Full documentation here:"  
[X Link](https://x.com/hwchase17/status/1695078260680622582)  2023-08-25T14:18Z 40.3K followers, [----] engagements


"Fine-tuning LLMs to generate output in a custom tone is a GREAT use case for LLMs Excited to announce a new set of loaders in @LangChainAI focused on loading data into a format to easily do exactly that Facebook WhatsApp Twitter Slack - all supported"  
[X Link](https://x.com/hwchase17/status/1695116901708624286)  2023-08-25T16:52Z 43.2K followers, 13K engagements


"Hard to imagine all this happened in the past two weeks 📰Latest LangChain Newsletter is out -an entire section on retrieval (lots of exciting stuff here) -what's new in Open Source & LangSmith -favorite blog posts and use-cases read and sign up here https://t.co/kmGI9Gm0TG https://t.co/nPmRwsoasg 📰Latest LangChain Newsletter is out -an entire section on retrieval (lots of exciting stuff here) -what's new in Open Source & LangSmith -favorite blog posts and use-cases read and sign up here https://t.co/kmGI9Gm0TG https://t.co/nPmRwsoasg"  
[X Link](https://x.com/hwchase17/status/1695135090689577151)  2023-08-25T18:04Z 62K followers, 45.5K engagements


"A great overview of what's going on under the hood in a RAG pipeline Thanks for writing this @czue"  
[X Link](https://x.com/hwchase17/status/1695175903519854842)  2023-08-25T20:46Z 45.8K followers, [----] engagements


"One of my favorite use cases for fine-tuning is getting a model to speak in a custom tone This is kinda hard to achieve with just prompting/few-shot Here's an example of fine-tuning a model to tweet like Elon Musk (compare it to what you get by just prompting)"  
[X Link](https://x.com/hwchase17/status/1695466818918650182)  2023-08-26T16:02Z 40.2K followers, 23.6K engagements


".@apify can do this pretty well Example here:"  
[X Link](https://x.com/hwchase17/status/1695470765343461756)  2023-08-26T16:18Z 45.8K followers, [----] engagements


"Part of what made our webinar on RAG so great in my mind was that we had an application builder (Pedro from Tavrn) on to talk about what he ACTUALLY does in practice If you are an app builder and want to join a future webinar - reach out Webinar:"  
[X Link](https://x.com/hwchase17/status/1695479215909724317)  2023-08-26T16:52Z 45.8K followers, [----] engagements


"We've started adding loaders to help load data in a format such that you can finetune (usually for a custom tone) So far we've got Facebook Messenger Slack Telegram WhatsApp Twitter (via @apify) Discord What else should we add"  
[X Link](https://x.com/hwchase17/status/1695490295914545626)  2023-08-26T17:36Z 45.8K followers, [---] engagements


"The cool thing about this is this can work with arbitrary ways of generating different embeddings for a document E.g. by splitting into smaller chunks creating a summary of the document generating hypothetical questions What other creative ways could we do this"  
[X Link](https://x.com/hwchase17/status/1695840345618358449)  2023-08-27T16:47Z 40.3K followers, 28.4K engagements


"I had a lot of fun hacking on this Will share more details on this during our webinar on fine-tuning LLMs to talk in your voice this Wednesday"  
[X Link](https://x.com/hwchase17/status/1695864893097103836)  2023-08-27T18:24Z 45.8K followers, 24.4K engagements


"Over the weekend I added a GMail ChatLoader integration to @LangChainAI to help anyone to finetune models on their emails 👀I evaluated it the old fashioned way - the eyeball test Here's how it did responding to various types of inquiries (VC customer integration @swyx)"  
[X Link](https://x.com/hwchase17/status/1696259958696985014)  2023-08-28T20:34Z 70.2K followers, 16.1K engagements


"Finally a VC Sadly I am heads down for the foreseeable future building cool things like this"  
[X Link](https://x.com/hwchase17/status/1696259964443246678)  2023-08-28T20:34Z 40.6K followers, [----] engagements


"Overall responses were pretty good but highlights the importance for grounding Check out the integration here: We're doing a webinar with @GregKamradt on this topic this Wednesday - register here:"  
[X Link](https://x.com/hwchase17/status/1696259966062248088)  2023-08-28T20:34Z 40.2K followers, [----] engagements


"@waydegilliam @OpenAI @LangChainAI you can use the .batch interface (see here: and set max_concurrency to determine the max concurrency (see example of that here:"  
[X Link](https://x.com/hwchase17/status/1696316483310350777)  2023-08-29T00:19Z 46K followers, [--] engagements


"This is really useful for finetuning models to replace GPT-4 at particular points in your system [--] Log all data flowing through your app to LangSmith [--] Export with some filter (all runs all runs with positive feedback etc) to a dataset [--] Finetune a model on this data"  
[X Link](https://x.com/hwchase17/status/1696938247031763164)  2023-08-30T17:29Z 43.6K followers, 24K engagements


"📈We launched monitoring in LangSmith today I've mainly been using for two use cases: 🕑 Track usage - are people actually using my app 👍/👎 Track feedback - are changes I introduced leading to a better user experience Can you spot when we upgraded our app to use an agent"  
[X Link](https://x.com/hwchase17/status/1696945105343754518)  2023-08-30T17:56Z 45.8K followers, 11.1K engagements


"Is @GregKamradt the best langchain webinar guest"  
[X Link](https://x.com/hwchase17/status/1697103336447840257)  2023-08-31T04:25Z 45.8K followers, [----] engagements


"This is great stuff from @MaksOpp and @deepsense_ai Let's you easily mask PII before passing it into the language model - and with LangChain Expression Language is very clear/easy how to plug that into your chain"  
[X Link](https://x.com/hwchase17/status/1697347969710989372)  2023-08-31T20:37Z 45.8K followers, [----] engagements


"🔊There's a lot of really cool data in audio files Big shout out to @patloeber and our friends at @AssemblyAI for adding a document loader This allows you to easily pull in and analyze (summarize question) audio data"  
[X Link](https://x.com/hwchase17/status/1697646748250911123)  2023-09-01T16:25Z 45.8K followers, 15K engagements


"📹Tabular Data Retrieval Webinar Retrieval of tabular data has a lot of subtleties. I'm very excited to deep dive on this with @the_cube_dev and @Delphi_HQ next week"  
[X Link](https://x.com/hwchase17/status/1697649265202409645)  2023-09-01T16:35Z 40.8K followers, 14.1K engagements


"My favorite webinars are always when we have real builders on. There's a lot of lessons you learn by actually putting applications into production That's why I'm so excited to have @Delphi_HQ on next week's webinar"  
[X Link](https://x.com/hwchase17/status/1698027255568507162)  2023-09-02T17:37Z 45.8K followers, [---] engagements


"Want to use graph database and vector stores together @neo4j now has a vector index. This means you can create vector representations for your nodes Read @tb_tomaz's excellent article on how this is integrated into LangChain:"  
[X Link](https://x.com/hwchase17/status/1698380334839005685)  2023-09-03T17:00Z 40.6K followers, 23.2K engagements


"Interesting deconstruction of agents Most agents are prompt + LLM + output parser in a while loop This makes that explicit by forcing users to implement the while loop - more customizability 🤖🪟 Agents Deconstructed Don't understand what's going on inside LangChain agents I made a new python package that exposes all the helper functions but lets YOU implement the "while" loop More understandable + more customizable https://t.co/Y4W4xURJhZ https://t.co/dX1rSQbRDn 🤖🪟 Agents Deconstructed Don't understand what's going on inside LangChain agents I made a new python package that exposes all the"  
[X Link](https://x.com/hwchase17/status/1698380893524762798)  2023-09-03T17:02Z 62K followers, 28K engagements


"Code Interpreter 🖥and Web Browsing🌐 were two of the first ChatGPT alts It's fun to see open source versions of both of them Web Browser by @MLBear2: Code Interpreter by @Shroominic:"  
[X Link](https://x.com/hwchase17/status/1698764614224556156)  2023-09-04T18:27Z 40.2K followers, 31.3K engagements


"One of the hardest parts of building LLM applications is making sure they have the right context @the_cube_dev is helping solve the context problem for tabular data We recently did an integration with them and we're going to talk about it in detail 👇"  
[X Link](https://x.com/hwchase17/status/1698767634119200810)  2023-09-04T18:39Z 45.8K followers, [----] engagements


"this is pretty interesting rather than a general purpose web-agent this gives each website owner the ability to create one specifically for their site bullish on this since successful agents generally have custom prompts + tools"  
[X Link](https://x.com/hwchase17/status/1698837114367164501)  2023-09-04T23:15Z 45.8K followers, 27.4K engagements


"A great deep dive into improving SQL QA by adding in a vectorstore tool and few shot examples @fpingham and @manuelsoria_ consistently add awesome stuff to the SQL implementation - excited to have them on the webinar tmrw to talk about this"  
[X Link](https://x.com/hwchase17/status/1699074620748652897)  2023-09-05T14:58Z 45.8K followers, [----] engagements


"🚀We re-launched LangChain Hub This time as a proper hub (no more janky GitHub repo) Liking playground uploading downloading - it's got it all Organization support coming soon"  
[X Link](https://x.com/hwchase17/status/1699117061782528353)  2023-09-05T17:47Z 42.5K followers, 28.2K engagements


"Really excited to be kicking off a hackathon with @streamlit One interesting project perhaps - using LLMs to create Streamlit apps Someone just uploaded a prompt to the Hub that takes a first stab at this"  
[X Link](https://x.com/hwchase17/status/1699141496140927034)  2023-09-05T19:24Z 40.3K followers, [----] engagements


"For other project ideas we wrote a little blog highlighting some here: There's also the AMAZING langchain streamlit github repo:"  
[X Link](https://x.com/hwchase17/status/1699141497747300634)  2023-09-05T19:24Z 42.5K followers, [----] engagements


"Great overview of RAGAS We also did a webinar with them [--] weeks ago - check it out for more insights of how to evaluate RAG systems"  
[X Link](https://x.com/hwchase17/status/1699156822643671427)  2023-09-05T20:25Z 45.8K followers, [----] engagements


"📈Trending on HN: prompt for comparing two mailing address. Now on LangChain Hub Complete with instructions on how to use LangChain Hub: Original Article: All credit to @benstein"  
[X Link](https://x.com/hwchase17/status/1699497030148841568)  2023-09-06T18:57Z 45.8K followers, 17K engagements


"If you have long chains written with LangChain Expression Language it can be really helpful to add human readable names to each component We've now got a cookbook for that"  
[X Link](https://x.com/hwchase17/status/1699832738369663354)  2023-09-07T17:11Z 45.8K followers, [----] engagements


"Excited to be visiting Berkeley next week to chat with the AI Entrepreneurs Group If you're on campus and want to attend - sign up here:"  
[X Link](https://x.com/hwchase17/status/1699916044071207060)  2023-09-07T22:42Z 43.2K followers, [----] engagements


"A great blog by @CobusGreylingZA on all the functionality the new langchain hub offers"  
[X Link](https://x.com/hwchase17/status/1700170212681953736)  2023-09-08T15:32Z 45.8K followers, [----] engagements


"A fantastic prompt for validating business ideas Added to LangChain Hub - play around with it here: I asked it to validate the idea of a "prompt hub" :)"  
[X Link](https://x.com/hwchase17/status/1700903960846979113)  2023-09-10T16:08Z 45.8K followers, 73.3K engagements


"Anyone want to help add a wechat chat loader to help finetune models to talk in your voice"  
[X Link](https://x.com/hwchase17/status/1701273898552270863)  2023-09-11T16:38Z 46K followers, [----] engagements


"Wanted to learn how to build LLM apps with audio @SmithaKolan from @AssemblyAI has you covered with an AWESOME new video"  
[X Link](https://x.com/hwchase17/status/1701286112445211032)  2023-09-11T17:26Z 45.8K followers, [----] engagements


"📹Superagent Deepdive Webinar📹 Superagent by @pelaseyed is a platform for building deploying and managing LLM agents (built on top of LangChain/LangSmith) Join us for a webinar this Thursday to do a deep dive on all the technology behind it"  
[X Link](https://x.com/hwchase17/status/1701640203285455316)  2023-09-12T16:53Z 45.8K followers, 42.7K engagements


"This is fantastic Also really cool to look at the prompts that are doing behind the scenes - all on the hub Blog generator: Tweet-from-text: Newsletter: Summarize:"  
[X Link](https://x.com/hwchase17/status/1701755423395914189)  2023-09-13T00:31Z 46K followers, 21.8K engagements


"The top prompt on the LangChain Hub is a 🥷Superagent prompt This prompt helps to power one of the leading agent platforms Explore the prompt here: And join @pelaseyed and I to discuss all things superagent related this Thursday https://www.crowdcast.io/c/ltmfb7aqb3ho https://smith.langchain.com/hub/homanp/superagent https://www.crowdcast.io/c/ltmfb7aqb3ho https://smith.langchain.com/hub/homanp/superagent"  
[X Link](https://x.com/hwchase17/status/1701785961934500091)  2023-09-13T02:32Z 62.8K followers, 15.6K engagements


"Great deep dive on evals by @databricks"  
[X Link](https://x.com/hwchase17/status/1702059912409354606)  2023-09-13T20:41Z 41K followers, [----] engagements


"If you like the LangChain "Talk to Notion" App but wish it had: ✂ Better splitting 🧠 Better memory 🎏 Newer @streamlit components 📓 Could be embedded back INTO Notion Then boy do I have a treat for you"  
[X Link](https://x.com/hwchase17/status/1702711873764762016)  2023-09-15T15:52Z 45.8K followers, [---] 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/anyuser/status/1703416519756267592)  2023-09-17T14:32Z 88.4K followers, 151.9K engagements


"When doing evaluation of RAG systems you often need to have example question/answer pairs This *highly* detailed prompt from @GitMaxd helps assist in that"  
[X Link](https://x.com/hwchase17/status/1703432371494879453)  2023-09-17T15:35Z 45.8K followers, 20.3K engagements


"Just made this work better with @AnthropicAI models Had to add two extra lines at the end: - Remind it to output longer summaries - Tell it to ONLY output JSON (otherwise claude is chatty)"  
[X Link](https://x.com/hwchase17/status/1703469056572195114)  2023-09-17T18:00Z 45.8K followers, 12.3K engagements


"🚀OpenAI InstructGPT [---] model OpenAI just quietly released a version of their [---] models available via the old completion endpoints This is nice because chat models are sometimes chatty which make them less ideal for agents where you want precise steerable outputs Example 👇"  
[X Link](https://x.com/anyuser/status/1703901472059265504)  2023-09-18T22:39Z 88.4K followers, 95K engagements


"To use in @LangChainAI it's very easy - just specify the new model gpt-3.5-turbo-instruct in the OpenAI class When using this model with agents I'd recommend using "old" prompts that are NOT in the ChatMessage format Example with @metaphorsystems below"  
[X Link](https://x.com/hwchase17/status/1703901473581740392)  2023-09-18T22:39Z 45.8K followers, [----] 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/anyuser/status/1704903312724693119)  2023-09-21T16:59Z 88.4K followers, 67.6K engagements


"If you want to jump right into it we've updated the "Getting Started" page for agents to go over all the individual components We then show how to create agents from these individual components Is a great resource to build up a solid base understanding"  
[X Link](https://x.com/hwchase17/status/1704903315182518736)  2023-09-21T17:00Z 42.5K followers, [----] engagements


"Tool rendering: LangChain makes it easy to create custom tools (with the tool decorator). The next question becomes how to pass them into the LLM We've add a few different methods and added a dedicated API reference section"  
[X Link](https://x.com/hwchase17/status/1704903320513437745)  2023-09-21T17:00Z 45.8K followers, [--] engagements


"Shout out to @shoggoth1313 for a lot of similar ideas His repo on "deconstructing" agents hit on a lot of similar components and we're excited to have worked with him to bring these into LangChain"  
[X Link](https://x.com/hwchase17/status/1704903334727962783)  2023-09-21T17:00Z 45.8K followers, [----] engagements


"This is maybe the most excited I've been for a webinar in a while @ShunyuYao12 has been at the forefront of a lot of this stuff really really excited to dive deep on how he sees this space"  
[X Link](https://x.com/hwchase17/status/1704909906212778481)  2023-09-21T17:26Z 45.8K followers, [----] engagements


"Very excited for this 🚀 We did a course with deeplearning that people loved. but that was only for Python ☹ @scrimba is awesome technology and educational content this should be a great partnership and resource"  
[X Link](https://x.com/hwchase17/status/1704949286033568086)  2023-09-21T20:02Z 45.8K followers, [----] engagements


"Sophia is one of the best educators - excited to see her giving this talk on langchain and RAG"  
[X Link](https://x.com/hwchase17/status/1704992117951705215)  2023-09-21T22:52Z 40.3K followers, [----] engagements


"Ex-@Kensho doing 🔥 in OSS/AI. I feel very lucky to have learned from all CharacterAI: @sam_shleifer @suno_ai_: Mikey Georg Martin New Computer: @sjwhitmore Magic Paper: @_mattneary Pebble: @whitej125 Trustfall: @PredragGruevski LangChain: @veryboldbagel Bearly: @pnegahdar"  
[X Link](https://x.com/hwchase17/status/1705229719191556402)  2023-09-22T14:37Z 40.3K followers, [----] engagements


"One of the most exciting agent launches last week was @startupandrew's launch of @Shortwave AI assistant 🤓Really excited to deep dive on this agent and how he built it with him next Tuesday Webinar link 👇"  
[X Link](https://x.com/hwchase17/status/1705603274852990986)  2023-09-23T15:21Z 45.8K followers, 25.4K engagements


"@steve_messina @HardKothari @LangChainAI Please do - support at langchain dot dev is best email"  
[X Link](https://x.com/hwchase17/status/1705661856915042331)  2023-09-23T19:14Z 40.3K followers, [--] engagements


"@GitMaxd @LangChainAI 1500+ langchain contributors is the greatest stat"  
[X Link](https://x.com/hwchase17/status/1705728213320900958)  2023-09-23T23:37Z 43.2K followers, [--] engagements


"A big part of building agents is making sure they have the right context Excited to talk with @startupandrew on how he uses AI to *find the right data* Register for the talk:"  
[X Link](https://x.com/hwchase17/status/1705965814950637802)  2023-09-24T15:22Z 45.8K followers, 12.2K engagements


"Want to deepdive into what this agent is capable of and how it accomplishes that 📹Join us on Tuesday: https://www.crowdcast.io/c/zeltomsizw2d Introducing the Shortwave AI Assistant Use AI to search your email history write emails in your voice schedule meetings and more. Available to everyone today for free: https://t.co/NaC9A1vViZ https://t.co/NuccbhzfQR https://www.crowdcast.io/c/zeltomsizw2d Introducing the Shortwave AI Assistant Use AI to search your email history write emails in your voice schedule meetings and more. Available to everyone today for free: https://t.co/NaC9A1vViZ"  
[X Link](https://x.com/hwchase17/status/1706328706124238932)  2023-09-25T15:24Z 61.9K followers, [----] engagements


"🤖AI Assistant webinar starting now We weren't going to do a webinar this week but then @startupandrew released Shortwave's AI Assistant It's going to be pretty casual - a walkthrough of the product and then questions. Really excited to deep dive"  
[X Link](https://x.com/hwchase17/status/1706714782177673607)  2023-09-26T16:58Z 45.8K followers, 13.9K engagements


"It's awesome to team up with @AssemblyAI for multimodal projects See their recent blog on using Assembly + @LangChainAI + @trychroma for Retrieval Augmented Generation over audio data:"  
[X Link](https://x.com/hwchase17/status/1707088826173448445)  2023-09-27T17:44Z 43.6K followers, [----] engagements


"This is a great integration @embedchain lets you easily create ChatGPT-like experiences over your data This integration with LangSmith will now let you easily debug and monitor those applications"  
[X Link](https://x.com/hwchase17/status/1707446474198642898)  2023-09-28T17:25Z 42.5K followers, [----] engagements


"@sarimrmalik @LangChainAI shoot us an email - hello at langchain dot dev"  
[X Link](https://x.com/hwchase17/status/1707548890998956303)  2023-09-29T00:12Z 45.8K followers, [--] engagements


"This is a great template for retrieval apps now with an interchange-able JS/Python backend"  
[X Link](https://x.com/hwchase17/status/1707890690271887758)  2023-09-29T22:50Z 45.8K followers, [----] engagements


"Awesome to see langsmith benefitting not just langchain users but also users of frameworks that build on top of langchain"  
[X Link](https://x.com/hwchase17/status/1708180033788100678)  2023-09-30T18:00Z 40.2K followers, [----] engagements


"This is SUPER easy to customize to the dataset you want to chat over In fact well be releasing more content next week on how to do exactly that 👀"  
[X Link](https://x.com/hwchase17/status/1708180258300764405)  2023-09-30T18:01Z 45.8K followers, 11.5K engagements


"Conversations as Directed Graphs with LangChain This is a really creative use case by Daniel Warfield"  
[X Link](https://x.com/hwchase17/status/1708904931598483706)  2023-10-02T18:01Z 40.3K followers, 12.3K engagements


"Lots of interest in how to handle PII when working with LLMs Good writeup by @fpingham on the topic including how to mask all inputs/outputs when sending data to LangSmith Doing a webinar on this next week - join us then https://www.crowdcast.io/c/6es3zwnv0ygy 🗃 Handle PII with LangChain Managing PII (personally identifiable information) data with LLMs can be especially tricky .@fpingham shares ideas for doing it safely 🤓learning the policies of LLM providers you rely on 🕵🏽♀anonymize data (w @presidio or @opaquesys) 🧽saving https://www.crowdcast.io/c/6es3zwnv0ygy 🗃 Handle PII with"  
[X Link](https://x.com/hwchase17/status/1709229524846776718)  2023-10-03T15:30Z 61.2K followers, 15K engagements


"Lots of people have asked for easy streaming of partial results for openai function calling (parsed into json) excited to easily support this"  
[X Link](https://x.com/hwchase17/status/1709691716067045548)  2023-10-04T22:07Z 45.8K followers, [----] engagements

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