I'm a big fan of sites like Exercism for learning new languages. Personally I get a lot more out of this practical approach than I would from reading a text book or watching lectures.

I'm wondering if there is something like that but for general data engineering problems, or if there's any interest?

For example, looking at the publicly available NYC citibike datasets (https://s3.amazonaws.com/tripdata/index.html), find the bike ID that travelled the longest distance in September of 2019.

There are many different approaches one could take to this problem, and it would be interesting to compare solutions based on performance, maintainability, etc...