1.2. The AWS Data Engineering Landscape

💡 First Principle: AWS offers dozens of data services, but they aren't random — they organize into clear categories aligned with the five pipeline stages. Think of your AWS toolkit like a well-organized tool belt: you don't carry every tool to every job, but you need to know which tool to grab for each situation.

The exam doesn't test whether you can name every AWS service. It tests whether you can choose the right one when given a scenario. A data engineer who understands why Glue exists (serverless ETL for data lakes) can answer questions about it even without memorizing every API parameter — because the "why" constrains the "when" and "how."

Unlike a solutions architect exam that covers networking, compute, and all of AWS broadly, this exam is laser-focused on the data path. You'll see the same 15–20 services appearing in nearly every question, with a handful of newer services (Bedrock, SageMaker Catalog, S3 Tables) making cameo appearances since the v1.1 update.