GraphStuff.FM: The Neo4j Graph Database Developer Podcast GenAI solutions with LangChain: Lance Martin on LLMs, agents, evals, and more!
35 snips
Feb 1, 2024 Expert Lance Martin explores GenAI solutions focusing on LLMs, agents, and evals. Topics include LangChain apps, RAG components, and optimizing retrieval processes. They also discuss tools like Streamlit, Olama, and the importance of Git for collaborative projects.
AI Snips
Chapters
Transcript
Episode notes
Lance's LangChain Journey
- Lance Martin discovered PaperQA, a RAG app for scientific literature, sparking his interest in LangChain.
- He built LexGPT, a RAG app for Lex Fridman's podcasts, which gained attention and led to his role at LangChain.
LLMs as Kernels
- LLMs are like the kernel of a new operating system, connecting to external resources like databases.
- RAG is a central component, linking LLMs to data sources for current information.
LangChain for LLM Development
- LangChain simplifies LLM app development by providing a framework to combine LLMs, databases, and embedding models.
- Its composability allows developers to easily swap components, like switching from Pinecone to a graph database.