GenAI Beyond Chat with RAG, Knowledge Graphs, and Python
summary
Hands‑on workshop that builds a working pipeline: embeddings and a Neo4j vector index for semantic search, a LangChain + OpenAI Python agent that chooses tools, and graph traversals that add directors/actors/genres as context. Demonstrations show RAG reducing hallucinations by passing JSON context, creating embeddings per plot, and a retriever that returns structured nodes; the movie agent finds titles (including a 2023 film missing from model training) and the demos all run live. Vector indexes and embeddings knowledge graph RAG movie agent
AI-written summary · about
chapters (9)
- 0:00Introduction to GenAI Beyond Chat
- 1:14Defining generative AI and LLMs
- 3:04Limitations of LLMs in applications
- 8:14Introduction to RAG and fixing hallucinations
- 11:45Graphs and vectors for data access
- 16:42Graph RAG with vectors and graphs
- 18:05Building a movie chatbot with agents
- 24:28Combining graph and vector search
- 25:30Q&A and resources
related talks
41mFrom RAGs to riches: Build an AI document interrogation app in 30 mins
01h 34mHands-on AI workshop: Graph RAG, Memory & Multimodal Agents
51mEvent Driven AI: Supercharging ChatGPT with RAG & LangStream
50mRetrieval Augmented Generation Beyond Toy Demos
58mDemystifying GenAI: Build a ChatGPT App with Vector Store
01h 13mBuilding Production-Ready AI Systems in 90 Minutes
29mAgentic GraphRAG: Multi-Agent Knowledge Graph Construction for Research Teams
48mGoing Beyond Prompt Engineering