Resource requests and limits, what a CPU number actually buys you, the metrics-server, and handing the replica count over to a Horizontal Pod Autoscaler.

Storage that outlives the pod: why container filesystems vanish, how PersistentVolumes and PersistentVolumeClaims split the job in two, an NFS server run with Docker Compose on EC2, and dynamic provisioning with DigitalOcean block storage.

ConfigMaps and Secrets: pull configuration out of your image, inject it as environment variables or mounted files, and understand exactly why a Secret is not private.

Ingress and ingress controllers: put many services behind a single load balancer, install nginx ingress on a real cluster, and route traffic by domain and path.

Services, hands-on in the cloud: expose pods with NodePort, put a real load balancer in front with LoadBalancer, and keep your database internal with ClusterIP.

Hands-on with pods, ReplicaSets, and Deployments: spin up a local multi-node cluster with Kind, create your first pod, and learn why you'll almost always write Deployments.

A plain-English starting point for Kubernetes: what a cluster actually is, what lives inside the master node, and what happens on the worker nodes where your apps run.

A node-by-node walkthrough of a GitHub PR review agent built as a LangGraph.js StateGraph — the shared state, every node's job, the parallel fan-out, the human gate, and the CLI that drives it all.