Containers stopped being optional some time ago, but most developers learn them by copying a Dockerfile from a blog post and hoping. This course teaches them from the mechanism up, quickly enough that you are productive on the first evening.
You begin with images and layers, and specifically with why layer order decides whether your build takes four seconds or four minutes. Then a Dockerfile written properly for a real web application: a multi-stage build that leaves compilers out of the final image, dependency installation cached separately from source code, a non-root user, a health check, and signal handling so the container actually stops when asked. You will measure the image before and after each improvement.
Compose comes next, and this is where daily life changes. One file that runs your application, a database, a cache and a background worker, with named volumes that persist data, networks that isolate services, environment configuration that differs between machines, and dependency ordering that waits for readiness rather than mere existence.
The last part is the operational reality: debugging inside a running container, reading logs, diagnosing a container that exits immediately, keeping images small and free of secrets, scanning for vulnerabilities, and a clear-eyed section on what changes when these containers run somewhere other than your laptop.