Platform-Engineering
- The Allow SCP that worked until it didn't
- I assumed GovCloud was AWS with a different region code. It took two weeks to prove me wrong.
- I debugged a Lambda timeout for 6 hours. The fix was 4 CLI commands.
- Your onboarding flow is your architecture's report card
- An orderly EKS and Kubeflow upgrade path
- Drift is an availability bug
- Kubeflow is a version matrix, not a version
- When a namespace owns your deployment