NODES AI 2026 - Agentic GraphRAG: Autonomous Knowledge Graph Construction and Adaptive Retrieval

In this session at NODES AI, Athulya Anil discusses "Agentic GraphRAG: Autonomous Knowledge Graph Construction and Adaptive Retrieval." The talk highlights the limitations of traditional GraphRAG systems, which often depend on manual schema design and rigid retrieval strategies. Athulya presents a multi-agent system that autonomously infers schemas, constructs knowledge graphs, and intelligently routes queries based on structure and risk signals. The speaker explores how specialized agents work together to manage entity extraction, resolve conflicts, and tackle multi-hop queries while optimizing retrieval strategies. By incorporating tools such as Neo4j, FAISS, and spaCy, attendees will gain insights into building adaptable retrieval systems that require minimal manual intervention and can evolve through systematic improvements.