Skip to content

Start here

These are the runbooks I use to learn platform engineering: AWS the way it’s run in production, Terraform, and Kubernetes on my own hardware. Each one is published only after I’ve run it end to end, with my real results and the things that broke.

  • Developers who know how to build software and want to learn how it’s deployed, secured, observed and recovered.
  • Anyone preparing for AWS Solutions Architect, Terraform Associate or CKA who wants hands-on labs, not just videos.

You should be comfortable in a terminal and with Git. No prior AWS or Kubernetes experience is assumed.

Track What you build Main tools
AWS A multi-account AWS setup, then the same platform as code, then a managed Kubernetes platform Organizations, IAM Identity Center, VPC, Terraform, EKS, Flux
Homelab A Kubernetes cluster on an old laptop and two Raspberry Pi 5s k3s, Ansible, Cilium, Argo CD, Prometheus

Both tracks run the same small Java (Spring Boot) service, so you can compare what each platform does for it.

  • AWS: most runbooks fit the Free Tier or cost a few dollars if you delete everything the same day. Each runbook states its cost and how to clean up. The first runbook sets budget alerts in every account — do it first.
  • Homelab: hardware you may already have; the Raspberry Pi route is about €450.
  • Every runbook shows the date I last ran it (Last verified). Cloud consoles change; if a button moved, the edit link at the bottom of each page is the fastest way to tell me.
  • These are learning labs, not production advice. Real systems need reviews, more accounts and more care.
  • You’re responsible for what runs in your AWS account. Read the Clean-up and cost section before you start a runbook, not after.
  1. Read How to use a runbook — two minutes.
  2. Start with A0 — An AWS organization nobody logs into as root. Everything else on the AWS track builds on it.