Happy Path Programming #113 Graph & RAG with Jennifer Reif
14 snips
Jul 21, 2025 Jennifer Reif, a Developer Advocate at Neo4j with a focus on Java and graph databases, dives into the world of integrating LLMs with data through Retrieval-Augmented Generation (RAG). She explains how RAG enhances responses from large language models by pulling in external data, while also discussing the differences between relational, NoSQL, and graph databases. Jennifer explores the advantages of graph databases in managing complex relationships and the future of AI in programming, emphasizing creativity in utilizing these powerful tools.
AI Snips
Chapters
Transcript
Episode notes
RAG Enhances LLM Responses
- Retrieval Augmented Generation (RAG) improves LLM answers by integrating external data.
- This addition retrieves relevant information and augments the LLM context for precise responses.
Relational Limits on Complex Queries
- Relational databases organize data in tables with rows and columns and use joins to combine data.
- Complex relationships degrade query performance because joins happen at query time, causing scalability issues.
Graph Databases Boost Query Efficiency
- Graph databases store data as nodes and relationships with physical connections on disk.
- This structure speeds up multi-hop queries and simplifies data refactoring and maintenance.