Customer Centric Observability at Scale Leveraging AIOps & OpenTelemetry
summary
Intuit builds an OpenTelemetry-based Real User Monitoring pipeline that ingests front-end spans and metrics into an operational data lake (S3 + ElasticSearch metadata), surfaces a unified anomaly score for alerting, and ties traces (Grafana Tempo) to logs (Splunk), metrics (Wavefront/Druid) and a service dependency graph. Goal is mean time to detect under five minutes and mean time to restore under 40 minutes with 99.99% availability. Platform auto-rolls back bad changes via Argo and uses adaptive span sampling to control cost.
AI-written summary · about
chapters (10)
- 0:00Intuit platform intro
- 4:35Reliability engineering foundations
- 6:00Customer-centric observability
- 8:07AI-driven observability goals
- 8:49Observability data pipeline
- 11:59Front-end performance monitoring
- 16:40Tracing platform architecture
- 26:04AI Ops anomaly detection
- 29:21Demo: dev portal & incidents
- 38:49Summary & Q&A
related talks
32mCustomer Centric Observability: How Intuit Reduced Time To Detect Customer Impact From 30+ Minutes To Under 3 Minutes.
42mUnified observability, AIOps, and incident response for AWS
28mScaling Observability for Real-Time Personalization Engines
27mAlert Fatigue to Action: Observability as the Backbone of Practical AIOps with OpenSearch
50mIntuit Journey To Unified Observability at Scale: Challenges, Be...
56mFull-stack observability and application monitoring with
13mNext-Gen Observability: reduce MTTR
12mMastering Observability in Distributed Systems