Cloud-Native Observability for Financial Systems
summary
Claims concrete outcomes: a 99.4% anomaly detection accuracy and 87% reduction in false alerts from a system built around graph-based topology analysis and machine‑learned behavioral baselines. Implementation uses OpenTelemetry and native integrations with AWS Azure GCP, handles 8 GB/day telemetry with adaptive compression and 42x faster processing, and cites Project Zenith savings (~$3.2M/year) and phased rollouts (assessment 2–3 weeks, pilot 3–4 weeks, scale 4–6 weeks).
AI-written summary · about
chapters (10)
- 0:00Introduction and Speaker Background
- 0:29The Challenge of Traditional Monitoring
- 1:59Current Monitoring Challenges
- 4:17Graph-Based Topology Analysis
- 6:54Behavioral Analytics Framework
- 9:08Performance Metrics and Business Impact
- 15:00Implementation Strategy
- 17:02Integration and Flexibility
- 18:37Getting Started and Next Steps
- 20:17Conclusion and Q&A
related talks
13mBuilding Financial Fortresses: Real-Time Fraud Detection
11mReducing exposure in the cloud
29mBuilding Resilient AI Platforms: Financial Fraud Detection
13mTransforming Financial Services with Real-Time Analytics
14mCloud-Native AI for Fraud Prevention with Bayesian Networks
13mBuild Real-Time Analytics Platforms for Financial Crime Detection
10mNeo4j Meets FinOps & GreenOps: Real-Time Graphs for Cloud Cost Intelligence
49mBuilding digital resilience with unified security and observability