Beginner books on the stuff I'm learning in public. Six chapters each, zero dumb questions — dangerous by the end.
Pods, deployments, services, ingress — from "what's a container" to running k3s on your homelab.
Desired-state infrastructure for people who click too much. Providers, state files, and real VM provisioning.
Move a real server to EC2 without the horror stories: instance anatomy, security groups, cloud-init, and a lockdown that sticks.
The 20% of REST, HTTP, auth, and curl that covers 80% of real work. Ends with you scripting an API like you mean it.
Git's mental model, Pages hosting, Actions, and the API — plus keeping your repos clean enough to show people.
Personas, skills, and memory as plain-text files that travel between AI tools — the AGENTS.md and Agent Skills standards, and the loop that makes agents smarter.
More books land as I learn things the hard way. Spotted a mistake? Hit the Feedback button — I read everything.
Short standalone guides from live builds — one real failure, diagnosed to ground truth, with the fix and the proof.
PBS as a VM, datastore on a passed-through drive, the token privilege-separation trap, prune/GC/verify — and the proof restore most guides skip.
Volumes stuck Degraded on a small cluster? The bundled StorageClass hardcodes 3 replicas and overrides your global setting. Diagnosis, durable fix, proof gate.