How I Recommend to Learn AWS Infrastructure
Less abstraction, more doing.
You can dive into AWS and run a lot of code pretty quickly. However, it becomes apparent how difficult it is to maintain everything built once you start running multiple services with dependencies. It’s similar to writing code until it needs to be maintainable, scalable, or worked on by various developers.
I’ll explain why I recommend building infrastructure on AWS without delving too deeply into high-level services.
Getting Started
I recommend using the UI to create services starting. The reason is that AWS automatically creates a lot for you, and you can focus on the service/thing that is trying to be created. For DS/ML folks, using SageMaker or Bedrock is nice compared to software engineers who will be interested in EC2, Lambda, etc.
Whatever it is, try running code on these services in the simplest way possible.
For example, creating a Lambda on the UI is simple. After some setup, you can point to the code uploaded to S3. You could also add a simple Cloudwatch event that kicks it off hourly.
Now, you can edit permissions to test another service without worrying about too many other details.
