How AI Agents Query Knowledge Graphs (a practical guide)

In the rapidly evolving field of AI, retrieval is only half the battle. The other half? Knowing how to move through your data - deterministically, intelligently, and at scale. This is a follow-up to our popular "How to Build a Knowledge Graph for AI" blog post. This time, we go deeper: instead of building the graph, we focus on navigating it - using an LLM to generate queries that traverse a rich, structured knowledge graph and return grounded, deterministic answers. We'll work with an e-commerce graph that combines relational data, embedded product descriptions, and customer reviews - showing exactly how an agentic system can reason across all of it. In this session you'll learn: - How to structure a knowledge graph that blends relational data with vector embeddings and graph edges - How LLMs generate graph queries to navigate complex, multi-model data - How to produce deterministic, citation-backed answers from a knowledge graph - How agentic memory fits into a full RAG pipeline Blog: https://surrealdb.com/blog/how-to-build-a-knowledge-graph-for-ai -- Chapters -- 0:00 Intro 0:48 Introduction and agenda 2:28 AI-first e-commerce use cases and outcomes 6:33 Building the context layer in SurrealDB 9:38 Knowledge graph schema and multimodal data 14:17 Unstructured ingestion and ETL pipeline 16:07 Agentic RAG and Text-to-SurrealQL tool 21:19 Simple file-based memory system 35:32 Observability and analytics-driven tuning 40:33 UI: JWT authentication and live queries 49:17 Q&A highlights and wrap-up --- Resources --- ☁️ Start free on Surreal Cloud: https://surrealdb.com/cloud 🎓 SurrealDB University: https://surrealdb.com/learn 📚 Docs: https://surrealdb.com/docs 🎩 Surrealist (GUI): https://surrealdb.com/surrealist ⭐️ GitHub: https://github.com/surrealdb/surrealdb --- What is SurrealDB? --- SurrealDB is an AI-native, multi-model database built in Rust that unifies document, graph, time-series, relational, geospatial and key-value data models into a single engine. Combined with powerful search and retrieval (vector, full-text, hybrid), real-time and event-driven capabilities, and Spectron - its persistent memory layer for AI agents - SurrealDB provides the complete context layer so that models run next to the data, context remains synchronised, and application logic stays simple and dependable. --- Connect --- 🐦 Twitter: https://twitter.com/surrealdb 👩‍💻 LinkedIn: https://www.linkedin.com/company/surrealdb/ 💬 Discord: https://discord.com/invite/surrealdb 🛍️ Merch: https://surrealdb.store/