LangServe: Deploying RAG Prototypes to Production

[@AI-Makerspace]

GitHub: github.com/AI-Maker-Space/LangServeDemo Slides: www.canva.com/design/DAF299aLBvc/TMRCvFv5bIaCMjLszbRfZA/edit?DAF299aLBvc&

[@andresrubio2015]

Greg, I am a fan of your Channel. Great job

[@AI-Makerspace]

Thanks @andresrubio2015 🤜🤛

[@NavjotMakkar]

What if I wish to add the conversation memory in this, how can we do that? For example, I build a Streamlit application for my rag chatbot so where does the memory object will be, will that be in my chain.py file or the Streamlit file? How will its work?

[@AI-Makerspace]

You'll want to add memory in your chain.py!

[@NavjotMakkar]

@@AI-Makerspacecurrently i am doing without memory and serving my langchain using langserve and then consuming it in my streamlit application. so what chain type should i use and memory type if i want to incorporate the chat history as well. Should i store chat message and send yhat as part of prompt to my api (langserve) or have that conversationmemory in my chain.

[@cavalierq]

just take the example from langchain and run it and talk as if its ground breaking. Greg - you should really just stop your BS and let Chris do more show and tell. wasted 1 hour Urgghh

[@AI-Makerspace]

Haha we agree that Chris is a legend with his demos! As a advanced practitioner, we'd love to hear from you about what you want him to tackle next!