O11y as Platform Contract: Instrumentation and CI/CD-Enforced SLOs
summary
Claims a working platform that delivers zero-code OpenTelemetry agents via base container images and sidecars, with CI/CD contract gating to enforce SLOs. Implementation instruments 200+ services across three clouds with 99.2% coverage, raises SLOs from ~23% to near 99.9%, cuts MTTR from 45 minutes to ~12 minutes (~73% faster), reduces manual rollbacks ~78%, and supports analysis in Data dog New Relic Grafana. Concrete architecture, measured KPIs, and rollout phases are described.
AI-written summary · about
chapters (12)
- 0:00Introduction and Overview
- 0:44Current State of Observability
- 1:57Challenges with Manual Instrumentation
- 4:12Introducing Platform Contract
- 7:53Zero Code Instrumentation Architecture
- 11:58Multi-Language and Multi-Cloud Support
- 13:20Observability Contracts and Enforcement
- 21:57Release Automation and Deployment Strategies
- 24:49Accelerated Incident Response
- 27:59Adoption Patterns and Industry Examples
- 32:03Blueprint for Success
- 35:10Conclusion and Q&A
related talks
45mPerformance-by-Design: Embedding Intelligent Scaling and Guardrails Into Platform Engineering
26mUsing OpenTelemetry to Improve Service Level Objectives
15mImportance of Observability in Platform Engineering
25mOpenTelemetry for Platform Engineers: Paving the Road To Self-Service Observability
44mAgentic Delivery, Guardrailed: AI in CI/CD & Platform Engineering (Open Source Patterns)
28mEnd-to-end CI/CD observability with open source
28mWhat's Next in Platform Engineering
31m20k Leagues Under Platform Engineering