Professional software developer since 1997, with much of that time spent helping developers build software that doesn't suck. A Typed Pure Functional Programming zealot who often compromises on his ideals to just get stuff done. Currently a Developer Advocate for AWS.
The Decision Layer: Context Graphs for Spring AI
– Room 8
Foredragsholdere
Ryan Knight
Ryan Knight is a Senior Partner Architect at Neo4j, where he works with AWS and Databricks to deliver public workshops and build Enterprise AI agents grounded in knowledge graphs. His work centers on how Knowledge Graphs can provide Enterprise agents the context for both data and decisions, grounded in a background in data engineering and distributed systems.
Prior to Neo4j, Ryan led Grand Cloud, a consultancy delivering data and generative AI solutions for clients ranging from startups to the Fortune 500. He has held senior architecture and advisory roles at Starbucks, Capital One, DataStax, and Lightbend.
Om sesjonen
Multi-agent Spring AI systems share data but not reasoning. The reasoning behind each agent's answer has no persistent form, and the next agent inherits the same data and none of the context the others built. This talk presents a Context-Aware Advisor built on the Spring AI Advisor API that acts as the Decision Layer. It intercepts every query, persists a structured decision trace to Neo4j, and resolves subsequent queries through graph-enriched decision search.
The graph is the advantage. Vector search finds decisions that look similar; graph traversal follows the relationships that make a prior decision applicable. A query that surfaces a conflict resolution traverses to the policy that authorized it, the exception that modified it, and every account that resolution has since governed. That structural context is complete grounding: not proximity to a document, but position in the decision record.
The talk ends with a complete multi-agent Spring AI architecture where agents share not just data, but the reasoning behind every decision that came before.
Passer for
Java and Spring developers building production AI applications. Spring AI exposure is helpful but not required. No graph database experience needed.
Samtidig i programmet
- Getting more out of Maven Room 1
- "35 Minutes Ago, You Had 20 Seconds to Comply": A Survival Guide to Days of Future Past Room 2
- Feature Management: Beyond Feature Flags Room 3
- Brukollapsen som bygde en bro mellom fag og IT Room 4
- Trust, But Verify: Skill-Driven Development for the Sceptical Java Developer Room 5
- Tilgjengelighet + utvikling = true Room 6
- Flyt i AI-ens tid: når det blir lettere å lage, men vanskeligere å bevege seg Room 7
- Rendering 3D shadows in the browser with three.js Room 6