Learn Kubernetes by testing one idea every day.
Short, focused experiments that you can run on any cluster in under 10 minutes.
Each day explores a single Kubernetes concept — from Deployments and NetworkPolicies to Operators, Helm, and Kustomize.
The goal is to learn by doing — and sometimes by breaking things.
Every experiment includes:
- A clear objective (what you’ll learn)
- A minimal YAML or CLI test
- A real outcome (success or failure)
- A short reflection (what it means)
You’ll develop muscle memory for daily Kubernetes operations and uncover how different parts of the system interact in practice.
daily/
├── day-001/
│ ├── README.md # Context, commands, and output
│ └── manifests/ # YAML files
├── day-002/
│ ├── README.md
│ └── manifests/
├── day-003/
│ └── …
└── …
Each directory stands alone — no dependencies between days.
You can start anywhere, but following in order helps build context.
Each day includes lightweight YAMLs safe to run in public sandboxes.
# Create a local cluster
kind create cluster --name k8s-lab
# Apply today’s manifests
kubectl apply -f day-001/manifests/
# Explore results
kubectl get all -A| Area | Examples |
|---|---|
| Core Concepts | Pods, Deployments, ReplicaSets, Services |
| Configuration | ConfigMaps, Secrets, Volumes, Probes |
| Scheduling | NodeSelectors, Affinity, Taints, Tolerations |
| Security | RBAC, NetworkPolicies, Kyverno, Falco |
| GitOps | ArgoCD, FluxCD, Helm, Kustomize |
| Operators | CRDs, Controllers, Custom APIs |
| Networking | CoreDNS, Ingress, Cilium, Service Mesh |
| Storage | PVCs, StatefulSets, CSI, Rook, Longhorn |
| Observability | Prometheus, Grafana, OpenTelemetry, Logs |
| Troubleshooting | kubectl tips, Metrics, Events, Debug pods |
| Scaling & Reliability | HPA, KEDA, Cluster Autoscaler |
| Policy & Governance | PSPs (legacy), Pod Security, Admission Control |
| Advanced | Multi-cluster setups, CAPI, Edge, GPU workloads |
Day 5 – Multi-Env Testing with Kustomize
Goal: Patch different deployment variants.
Command: kubectl apply -k kustomization-examples/env1.
Observation: env1-my-app deployed with its own labels.
Lesson: Overlays make environment management trivial.
- Builds intuition faster than passive reading
- Reinforces core commands through repetition
- Encourages “fail fast” learning habits
- Helps you see Kubernetes as a system, not a checklist
Get new experiments delivered daily: 👉 Join the newsletter
Or follow along on Twitter/X: @KubeSkills
Want to add your own “day”?
- Fork the repo
- Create a new folder dayXX
- Add: • README.md (context, commands, lessons) • manifests/ (YAML or Helm values)
- Open a pull request
Your test idea might become part of the next release of the newsletter.
MIT © 2025 KubeSkills
“Fail fast. Learn faster. Ship knowledge daily.”