Knowledge Graphs for AI Agents (A Practical Guide)

👉 RAG systems hit a wall as they scale: dense vector neighbourhoods produce context clash, hallucinations creep into answers, and high-dimensional embeddings stop being enough on their own. This session uses real-world examples — e-commerce search, customer support, agentic chatbots — to analyse where vector-only RAG breaks down, and the operational cost of stitching together separate vector, graph, and full-text stores. Martin (Solutions Engineer at SurrealDB) demonstrates how knowledge graphs ground LLM responses by adding deterministic structure on top of vectors, walking through the full ETL pipeline — parsing, chunking, embedding, entity and relationship extraction, deduplication, ontology alignment, and loading — and shows how a single SurrealQL query can combine vector search, multi-hop graph traversal, and full-text search to serve agentic RAG from one live dataset, avoiding the architectural "duct tape" of multi-database glue code. 🖥️ Explore other SurrealDB webinars: https://sdb.li/4sOML5J 📃 Read the guide: https://sdb.li/3R3p6RG --- Chapters --- 00:00 Intro 03:55 What is a Knowledge Graph? 11:19 Why vector-only RAG breaks 21:48 Q&A: How GraphRAG actually works 25:13 Building a KG: Extract, Transform, Load 43:23 Demo: exploring a KG in Surrealist 46:49 Agentic RAG & tool calling 50:38 One query: vector + graph in SurrealQL 58:31 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/