Summary

Keywords

neo4j graph databases graphs LLM retrieval augmented generation RAG vector database vector search GenAI AI Artificial intelligence nodes25

Full Transcript

Multi-agent AI systems often operate as mysterious ""black boxes"" making it difficult for developers and users to understand decision-making processes, trace knowledge sources, or debug unexpected behaviours. This session demonstrates how to build transparent and explainable multi-agent systems using Neo4j's capabilities and visualisation. You'll learn how to: • Design multi-agent architectures • Implement Neo4j GraphRAG to provide transparent knowledge retrieval and reasoning paths • Visualise agent decision-making processes through graph relationships • Track knowledge provenance from source documents to final answers • Debug and optimise agent interactions using graph-based insights • Build real-time dashboards showing agent collaboration and knowledge flow This will be a live demo of a complete multi-agent system featuring specialised agents (research, analysis, code generation) coordinated by a supervisor, with Neo4j GraphRAG providing transparent knowledge access. Attendees will see live demonstrations of agents collaborating, making decisions, and presenting their reasoning through interactive graph visualisations. Walk away with practical patterns for building trustworthy, explainable AI systems that developers and users can understand, debug, and improve. Speaker: Morgan Senechal Resources: Get Started with Aura - https://bit.ly/3LOLrjh Deployment Center - https://bit.ly/4jOelM3 Ground AI Systems and Agents with Neo4j - https://bit.ly/4oVsnyb #nodes2025 #neo4j #graphdatabase #graphrag #knowledgegraph