Cloud Native Development Without the Toil
summary
Argues safety-first delivery unlocks speed and shows concrete practices: use artifact syncing (Skaffold or Tilt) to push class-level changes to a remote cluster instead of full image rebuilds, use Telepresence to intercept traffic and debug locally, and use Argo Rollouts to declaratively run a 20% canary then pause and auto-analyze metrics for rollback. Recommends buildpacks, Helm, Prometheus/Jaeger, GitOps (Argo CD/Flux) and paved‑road platforms to remove toil.
AI-written summary · about
chapters (22)
- 0:00Intro
- 0:26tl;dr
- 2:30Why we're all here: Happy users
- 3:09Continuous delivery of value
- 4:342 modern software trends for speed & safety
- 5:40Trend 2: Kubernetes
- 7:33The Kubernetes migration journey(s)
- 10:52Safety requires multiple strategies
- 11:53Safety requires the cloud
- 13:26Safety needs to be part of your workflow
- 18:44Patterns, practices & tooling
- 20:25Skaffold: "Source to k8s" CLI tool
- 21:57Pattern: Dev environment bridging/extension
- 22:41Telepresence: Local-to-remote bridge
- 24:41Pattern: GitOps
- 26:48Argo CD: GitOps for k8s
- 28:10GitOps in a nutshell
- 28:54Argo rollouts: k8s progressive delivery controller
- 30:49Putting it all together: Safety & speed
- 32:23What about the platform & the people?
- 33:37Conclusion
- 34:26Outro
related talks
39mCI/CD, APIs, and Scaling: What Every Cloud Native Developer Needs To Kn...
32mCloud Native Progressive Delivery
46mIn Search of the Perfect Cloud Native Developer Experience
52mCloud Native Developer Experience in 2025
36mCloud-Native Dev Tools: Bringing the cloud back to earth
42mCloud Native Applications
30mApp Delivery in Cloud Native: Where are We? What's Next?
44mCloud Native Applications with Remote Development Environments