Enterprise discipline, small-team size.
I'm an SRE working in enterprise finance. My day job is the heavyweight version of Kubernetes: platform teams, change processes, audits, regulators — an environment where skipping a step has a visible price tag.
Kestrion is the opposite exercise. Most companies running Kubernetes have three to ten engineers and no platform team at all. They inherited a cluster, or migrated to one, and now carry enterprise-shaped operational load with startup-shaped headcount. The advice they find online either assumes a platform team they don't have or sells them another tool they don't need.
So these notes document one question, over and over:
What is the minimum discipline a small team needs to run Kubernetes safely — and what enterprise baggage should they refuse to copy?
The habits scale down: named ownership, everything in Git, tested restores, scheduled upgrades, a security baseline you could show an auditor. The machinery doesn't: service meshes by default, multi-cluster architectures, internal developer portals for five engineers. Copy the discipline; skip the machinery.
What's here
- The Production Readiness Scorecard — 50 questions, scored 0–100 in your browser. The fastest honest picture of where a cluster stands.
- Field notes — long-form pieces on cost, deploy safety, triage, security and audits.
- The newsletter — one field-tested pattern per week. No vendor pitches.
What this is not
Nothing here is sponsored, and no tool vendor is behind it. I'm collecting patterns — from research, from conversations with small teams, and from calibrating the scorecard against real clusters. If your team runs Kubernetes with fewer than fifteen engineers and something on this site sounds familiar, I'd genuinely like to compare notes: subscribe and reply to any issue.
Also from Kestrion
cka.kestrion.dev — a free CKA (Certified Kubernetes Administrator) study course, in Spanish.