Kubernetes
Learn Kubernetes from pod to production. Deployments, services, ingress, Helm, scaling, RBAC, and observability.
14 lessons, in order
- 125 min
What is Kubernetes
Welcome. If you have used Docker once or twice and now you keep hearing the word Kubernetes without quite knowing what it is, you are in the right...
Why orchestrators existControl plane vs nodesKubectlCluster-infoManifests - 230 min
Your First Pod
In the last lesson you talked to a cluster, but you never asked it to run anything. That is the gap this lesson closes. Kubernetes does not run a...
Pod manifestKubectl applyKubectl getKubectl logsKubectl execKubectl delete - 335 min
Deployments and ReplicaSets
You saw it yourself at the end of the last lesson: delete a bare Pod, or let it crash, and nothing brings it back. That is fine for a five minute...
Deployment manifestReplicasReplicaSetRolling updatesRollout statusRollout undo - 435 min
Services
You have a Deployment keeping three Pods alive, which solves the problem of Pods dying and not coming back. It does not solve a second problem that...
ClusterIPNodePortLoadBalancerSelector matching labelsKubectl exposePort-forward - 535 min
Probes and Self-Healing
A process that has crashed is easy to deal with. It is gone, the operating system noticed, and Kubernetes starts a new one. The hard case is the...
Liveness probeReadiness probeStartup probeHttp probe specRestart policyWhat happens on failure - 635 min
ConfigMaps and Secrets
Right now, if your container needs a setting, such as a log level or a database username, the only place to put it is baked into the image itself....
ConfigMap manifestEnvFromMount as fileSecret base64Kubectl create secret genericNever commit real secrets - 735 min
Volumes and Persistent Storage
Everything you have deployed so far has been amnesiac on purpose. A Pod starts, does its job, and when it dies the filesystem it was writing to dies...
EmptyDirPersistentVolumeClaimStorageClassPVC manifestMount in a Deployment - 825 min
Namespaces and Labels
A cluster with six Pods in it is easy. You run kubectl get pods, you see everything, and you know what each one is. A cluster with six hundred Pods...
Kubectl create namespace-n flagDefault vs kube-systemLabelsSelectorsKubectl get -l - 935 min
Ingress
You can already get traffic into the cluster. NodePort opens a high-numbered port on every node; LoadBalancer asks your cloud provider for a public...
Ingress controllerIngress manifestHost and path rulesIngressClassNameTLS via cert-manager - 1035 min
Helm Basics
Write your fifth web application for Kubernetes and you will notice something dispiriting: you keep typing the same YAML. A Deployment, a Service, a...
Chart structureValues.yamlHelm installHelm upgradeHelm templateWhen to use Helm vs raw manifests - 1130 min
Scaling and HPA
Picture a supermarket with ten checkout lanes but only two cashiers on duty. At nine in the morning that is plenty. At six in the evening the queue...
Kubectl scaleMetrics-server requirementHorizontalPodAutoscaler manifestCPU targetVPA mention - 1235 min
RBAC
Every command you have run in this course went out as you, the cluster administrator, and the cluster said yes to all of them. That is fine while you...
ServiceAccountRole vs ClusterRoleRoleBindingLeast privilegeVerbs and resources - 1330 min
Observability
At three in the morning something is wrong. Users are seeing errors. You have a cluster with forty Pods in it, and the only honest description of...
Stdout-as-logsKubectl logsPrometheus + grafanaOpenTelemetry mentionWhere to start - 1460 min
Capstone: Deploy a Stack
Every lesson so far has taught one object at a time, and every one of them was demonstrated on its own. That is how you learn a thing, and it is not...
Multi-service DeploymentConfigMap for configSecret for db passwordPVC for db dataIngress for the APILiveness probe+1
Learn Kubernetes in your language
Every lesson, quiz, and tutor reply renders in your language. The AI tutor matches the language you ask in.
Roles that use Kubernetes
See how hiring teams interview engineers on Kubernetes, in production-like environments.
Platform Engineer
Hire platform engineers who reduce friction for every team. Real golden-path templates, real internal CLIs, real service onboarding flows. We record every step.
See the interviewDevOps Engineer
Hire DevOps engineers who actually ship. Run live or take-home interviews in production-like Linux environments with Docker, Kubernetes, Terraform and CI/CD. See how they debug, automate, and work alongside AI.
See the interviewSite Reliability Engineer (SRE)
Hire SREs who actually run production. Live workspaces with metrics, logs, traces and a service that is starting to misbehave. We record every keystroke.
See the interview