← Back to the course home
🗓️ The study plan — 26 lessons, 9 weeks
The whole course in sequence, ~3 sessions a week, each session = read one
lesson (~15 min) + do its lab (~25 min). Tick lessons off as you go — progress is saved in
this browser. Don't rush: the labs are where it sticks.
🐢 Steady (recommended): 3 sessions/week → done in ~9 weeks. Fits around a job.
🐇 Fast track: ~2 hours/day → done in ~2½ weeks. Good between jobs or on break.
💰 Cost: weeks 1–8 are laptop-only (free). Only the optional capstone touches AWS — destroy same day.
WEEK 0 🎒 Foundations check (optional · 1–2 evenings)
Only if terminal/git/YAML feel shaky — run the self-checks on the
Before-You-Start checklist. All green? Skip straight to Week 1.
Install Docker Desktop + kubectl + minikube now either way.
WEEK 1 🍱 See it run — the core loop
From "what's a container" to a self-healing, reachable app on your laptop.
🏁 Checkpoint: delete a pod on purpose — then explain, out loud, who brings it back (L03) and how traffic still finds it (L04).
WEEK 2 🔑 Configure it properly
Rooms, settings, secrets, health and portions — the difference between "runs" and "runs right".
🏁 Checkpoint: liveness vs readiness in one sentence each, from memory — and cause (then explain) one OOMKill.
WEEK 3 🚌 Traffic & change
Scale under load, route from outside, and ship new versions with zero downtime.
🏁 Checkpoint: ship a broken image, watch readiness protect you, roll back in one command. If that felt boring — perfect, that's the point.
WEEK 4 📚 State & internals
What survives, and what's actually happening underneath.
+ dessert: the under-the-hood diagrams and the 6-min video tour.
🏁 Checkpoint: narrate kubectl apply end-to-end from memory — principal, register, timetable-maker, class teacher. Halfway there! 🎉
WEEK 5 🤖 Ship it — the bonus band
Robots deploy it; buildings and ladders keep it alive.
💡 Detour worth taking now (or after week 9): the sister courses —
Docker & ECR ·
AWS: IAM & EC2 ·
ArgoCD in depth.
🏁 Checkpoint: hand-edit the cluster and watch ArgoCD revert you. Recite the ladder: HPA adds pods → autoscaler adds desks → spread across buildings.
WEEK 6 🩺 Operate I — when things break
Part 3 begins: triage, permissions, and scheduled work.
🏁 Checkpoint: diagnose all three sick pods from L16's infirmary without peeking — and apply the real backup CronJob.
WEEK 7 🚫 Operate II — locks & seats
Network rules, seating rules, and the pods on every floor.
🏁 Checkpoint: the spy pod gets blocked; taint vs affinity explained in school words (push from desks, pull from kids).
WEEK 8 🏗️ Operate III — state, pressure, renovation
The advanced trio every on-call engineer eventually meets.
🏁 Checkpoint: kill postgres-0 and find the data intact; cordon a node and explain what a drain would wait for (the PDB!).
WEEK 9 🎓 Graduate — extend, observe, build
The last two lessons, then prove it all with a capstone.
🏆 Capstone (pick one):
💻 Zero-cost: on minikube, deploy everything — both apps, HPA, NetworkPolicies, the backup CronJob, postgres-0, and the Prometheus stack — then break three things and fix them using only L16's drill.
☁️ Real AWS (~$5 if destroyed same day): terraform apply → deploy via the pipeline → see it live behind the ALB → terraform destroy before dinner. 🧹
🎓 26 / 26 — you did the whole school!
Next: the ArgoCD course for deployment mastery,
or go build something real — you're ready.