Keynote: Think Faster: Graphs and AI for Real-Time Crisis Resolution
summary
A production ETL pipeline samples network traffic 2400 flow records every 5 seconds and builds live Neo4j graphs that correlate ExtraHop flow data, Splunk firewall logs and F5 load‑balancer events across millions of nodes to surface undocumented SMTP, SSH and backup jobs. The team extends the window to 12 days and reaches 6% sampling = operational equivalence, then layers Neo4j Aura and Bloom with Anthropic Bedrock models (Claude/Sonnet 4.5) to diagnose 500 errors and propose or execute fixes via an agent orchestration loop.
AI-written summary · about
chapters (6)
- 0:00Introduction and background
- 3:38Graph technology for business separation
- 10:26Integrating AI with business data
- 17:53Real-time problem resolution with graphs
- 21:49Future of AI-driven software management
- 24:00Conclusion and Q&A
related talks
30mContext Engineering: Connecting the Dots with Graphs
30mPerformant GraphRAG Improving Neo4j to Drive Autonomous AI ...
12mLeveraging Neo4j Graph Data and AI Across the App Development and Deployment Lifecycle
14mUsing Graphs to Fine-Tune Agents
25mLakehouse + Graph: Delta Lake, Unity Catalog and Neo4j for Governed AI Pipelines
21mFrom Data to Knowledge to Action - The Graph Intelligence Platform
59mOpening Keynote (Live)
51mMemory Masterclass: Make Your AI Agents Remember What They Do!