Leveraging Knowledge Graphs for RAG: A Smarter Approach to Contextual AI Applications

This 38-minute conference talk presented by David vonThenen from DigitalOcean at All Things Open AI 2025 explores how knowledge graphs can revolutionize retrieval-augmented generation (RAG) systems for more effective contextual AI applications. Discover why graph databases offer compelling advantages over traditional vector databases when handling complex data relationships in AI systems. Learn practical skills for building RAG systems with graph databases, including techniques for data querying and insights retrieval from both structured and unstructured sources. The presentation includes concrete examples of when to choose graph-based approaches for superior business outcomes, advanced implementation techniques, and culminates in a live demonstration showcasing the key concepts discussed. Ideal for developers and AI practitioners looking to enhance LLM performance through better contextual understanding and data relationship management. Slides are available on SlideShare for reference after the talk.