Beyond the theory, if you are having trouble getting your hands dirty, here is the path I had taken for data engineering:
(It is pretty hands on.)
After you are done with the initial learning, find an academic machine learning discord or something similar. There will always be people there who will be very happy to find someone to clean their data. It's a great way of getting hands on with data engineering.
System design is best learnt through fires.
A good angle of attack is: pick a certification like AWS SAA or equivalent(AWS, azure, Google cloud. Doesn't really matter. Just pick a mid level certification).
Then do the labs. They will quickly point out the holes in your knowledge/understanding. The free tier will take care of your needs, and cloud providers most of the time forgive the first surprise bill(it happens to everyone).
Soon data engineering and systems design merge anyways. This path is like a cheat-code for forcing convergence. Otherwise, the mind will quickly forget most theory.
---
Adding to this:
When you get hands on, you will find how the mind lies to you about how much you know.
Mind maps are a nice way to reliably detect holes in understanding. Sit down with paper & pen once a week or so, and make a big map of everything you know.
Data engineering and systems design are incredibly close. After some time, they completely mesh together.
And from personal experience, the knowledge-volume to become practically potent is actually smaller from data engineering side (than from normal software development side).
I've found it easier to get data gigs and simultaneously strengthen the knowledge.
(It is pretty hands on.)
After you are done with the initial learning, find an academic machine learning discord or something similar. There will always be people there who will be very happy to find someone to clean their data. It's a great way of getting hands on with data engineering.
System design is best learnt through fires.
A good angle of attack is: pick a certification like AWS SAA or equivalent(AWS, azure, Google cloud. Doesn't really matter. Just pick a mid level certification). Then do the labs. They will quickly point out the holes in your knowledge/understanding. The free tier will take care of your needs, and cloud providers most of the time forgive the first surprise bill(it happens to everyone).
Soon data engineering and systems design merge anyways. This path is like a cheat-code for forcing convergence. Otherwise, the mind will quickly forget most theory.
---
Adding to this:
When you get hands on, you will find how the mind lies to you about how much you know. Mind maps are a nice way to reliably detect holes in understanding. Sit down with paper & pen once a week or so, and make a big map of everything you know.