20k Leagues Under Platform Engineering
summary
Three-year platform-building story that turns manual on-premises deploys taking four to five hours into a self-service IDP with Kubernetes on EKS, GitLab CI, artifact repository and GitOps pipelines using ArgoCD and FluxCD (Argos chosen for the logo). Key incidents: SonarQube cloud misconfiguration exposing code, late addition of observability (New Relic → recommendation: OpenTelemetry/Grafana), a staged rollout with one pilot then mandatory adoption once ~70% of products move to the platform, and eventual lift-and-shift to AWS Milan.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:53Manual chaos to platform vision
- 3:16Automation and standardization
- 8:57Aligning platform and business
- 12:31Kubernetes and CI/CD tools
- 15:47Argo vs Flux security gaps
- 17:56Observability and developer experience
- 19:20Platform as a product
- 21:50Lessons learned and evolution
- 24:00Final takeaways
- 27:54Q&A
related talks
28mWhat's Next in Platform Engineering
31mPlatform Engineering's Inferno
40mAI-Augmented DevOps with Platform Engineering
44mPlatform Engineering in action
27mPlatform Engineering with the Argo Ecosystem: The Elastic Story
54mFrom Code to Customer
33mBlueprints of Innovation: Engineering Paved Paths for a User-friendly Developer Platform
02h 47mDes silos au Platform Engineering en passant par le DevOps : adopter GitOps et aller au-d...