The Graph Layer Behind NASA’s Breakthroughs | Michael Hunger

In this episode of The AI Native Dev podcast, Simon Maple interviews Michael Hunger, now in Product Innovation at Neo4j, about graph databases and their applications in AI. They discuss the advantages of graph databases over relational databases, especially in representing relationships and context, and how graph databases can enhance Retrieval-Augmented Generation (RAG) systems by providing richer context and explainability. They also delve into Model Component Protocol (MCP) and its potential for improving developer productivity by integrating infrastructure APIs into the development environment, while also addressing security concerns related to MCP servers, emphasizing the importance of trusted sources and sandboxing. The conversation concludes with practical advice on utilizing MCP servers and building custom ones for better understanding and control.