AI Business LensTHE BUSINESS OF AI, FOR PEOPLE WHO TEACH IT OR LEARN FROM IT
InfoQ · September 12, 2026

Presentation: From Retrieval to Reasoning: Building Production-Ready Agentic AI Systems with Knowledge Graphs

Computer ScienceData ScienceInformation Systems
THE AI ANGLE
Operating as autonomous agents taking over software development and reasoning tasks

Cassie Shum outlines an architectural shift from using knowledge graphs merely for retrieval to employing them as the structural reasoning foundation for agentic AI systems. She presents patterns such as decision provenance, context bundling, and agent visibility to resolve issues with long-context LLM prompting like inconsistency and lack of auditability. For computing and information systems educators, this highlights how foundational data modeling techniques remain vital for governing and stabilizing autonomous AI in software engineering.

THE TEACHING ANGLE
Students can explore the architectural trade-offs between relying on massive LLM prompt windows versus using structured knowledge graphs to ensure consistency, accountability, and multi-domain dependency tracking in autonomous systems.

Read the original at infoq.com   Generate teaching or study materials

Instructors get discussion guides, assignments, and mini-cases. Students and readers get a plain summary, class prep, and an exercise. All built from the full article. Three are free with an account.

More in Computer Science