When I first started learning AWS, my natural approach was simple:
Learn what S3 does.
Then Lambda.
Then EC2.
Then DynamoDB.
But I quickly realized that knowing what individual services do doesn’t automatically mean I know how to build something with them.
So I’ve started changing the way I learn AWS.
Instead of asking:
“What does this service do?”
I’m trying to ask:
“What problem would make me need this service?”
For example, learning that Amazon S3 provides object storage is useful.
But it becomes much clearer when I imagine an actual system:
User uploads a file → S3 stores it → Lambda processes it → results are saved → CloudWatch helps monitor what happened.
Suddenly, I'm not memorizing four AWS services.
I'm understanding how they work together.
I’m trying to use the same approach while learning AI and data services.
Instead of simply memorizing Amazon Bedrock, SageMaker AI, Glue, Athena, or other services, I want to understand where each one fits inside a real architecture.
My current learning rule is:
Concept → Problem → Architecture → Hands-on practice
If I learn something new, I try to connect it to a problem and eventually build a small version of it.
It takes a little more effort than simply finishing courses, but I’m already finding that the concepts stick much better.
Still early in the journey, but this feels like a much better way to learn cloud.
Loading article