Presentation: From Retrieval to Reasoning — Building Production Agentic AI Systems with Knowledge Graphs
Cassie Shum argues that knowledge graphs are a foundational component for agentic AI systems and outlines four practical architectural patterns — context bundling, decision provenance, code as truth, and agent visibility. She also demonstrates an engineering harness built on a knowledge graph aimed at streamlining feedback loops, optimizing token usage, and improving system reliability.