1.2. The Azure Back-End Toolkit
💡 First Principle: Azure services arrange themselves along a single axis: how much of the underlying machinery you control versus how much the platform manages for you. More control means more operational burden; more abstraction means more constraints. Nearly every "which service should you use?" question on this exam is asking you to locate the requirement on this axis.
This section matters because service selection is the single most common AI-200 question pattern. A scenario names constraints — "minimize operational overhead," "the team needs kubectl access," "scale to zero when idle" — and each constraint is a signpost pointing at exactly one spot on the control-abstraction axis. Learn the signposts and these questions become free points.
The mental model: think of it like transportation. AKS is owning a car (total control, you do maintenance), Container Apps is a rental (you drive, someone else maintains), App Service is a taxi (tell it where to go), and Functions is a moving walkway (step on, it runs only while you're on it). No option is "best" — each is best for a stated constraint, and the exam always states the constraint.
⚠️ Common Misconception: Developers pick the service they know rather than the one the requirement names. The exam punishes this: a scenario saying "the operations team must apply existing Kubernetes manifests" has already eliminated everything except AKS, no matter how attractive Container Apps sounds. Train yourself to hunt the constraint keyword first.
