Foundations & Architecture
Why orchestration, the control plane and node components, kubectl, declarative YAML, and running your first local cluster. Start →
Kubernetes — From Zero to Hero
A standalone, developer-focused course on Kubernetes. From the control plane and the declarative model to Pods, Services, storage, autoscaling, RBAC, and shipping to production with Helm and GitOps — taught through real manifests, kubectl, and diagrams.
What you’ll learn
Section titled “What you’ll learn”Workloads
Pods, Deployments and rolling updates, StatefulSets and DaemonSets, and batch work with Jobs and CronJobs. Deploy →
Networking
Services and kube-proxy, cluster DNS, Ingress and the Gateway API, and locking down traffic with NetworkPolicy. Connect →
Storage & Configuration
Volumes, PersistentVolumes/Claims and StorageClasses, ConfigMaps and Secrets, and namespaces with resource quotas. Configure →
Scheduling & Scaling
Requests/limits and the scheduler, affinity/taints/tolerations, the HorizontalPodAutoscaler, and disruption budgets. Scale →
Observability & Security
Liveness/readiness/startup probes, logs and metrics, RBAC and ServiceAccounts, and Pod Security Standards. Secure →
Production & Ecosystem
Packaging with Helm, extending the API with Operators and CRDs, GitOps, and a production checklist with the TypeScript API client. Ship →
How Kubernetes Works
The API server and etcd watch, controllers and informers, the scheduler framework, the admission chain, and Server-Side Apply internals — proven on a real cluster and cited from the 1.37 source. Look inside →
Reading AI-generated Kubernetes
A 22-bug catalog of what AI-generated YAML gets wrong, a review checklist, verification tools with a static schema checker, and three worked reviews — every bug proven. Review →
Glossary
89 terms from Pod to ValidatingAdmissionPolicy, with the Kubernetes version each became GA or was removed. Look up →
How this course works
Section titled “How this course works”The model first, then the manifest
Every topic starts from a model — the reconciliation loop, the Pod, the Service, the scheduler — so Kubernetes’s behavior becomes mechanism you understand, not YAML you copy-paste.
Verified against current API groups
Written against the stable API groups in wide production use today — apps/v1, batch/v1, networking.k8s.io/v1, autoscaling/v2, policy/v1 — and checked against the official docs via context7. Last verified against Kubernetes 1.37 (kubectl 1.36 client, kind 0.33), Helm 4.3, and current Gateway API v1, as of September 2026.
Guarantees, not just YAML
Not “here’s a Deployment” but “here’s what a rolling update actually guarantees, and what a missing readiness probe or PodDisruptionBudget can still let go wrong.” You’ll reason about a cluster with confidence.
Diagrams that explain
The control plane, the reconciliation loop, Service routing, and the scheduling pipeline as theme-aware Mermaid diagrams. Bilingual (English / ไทย) with saved quiz progress.