[Tainnor]

> try a few easy leetcode problems in Lean with a proof of correctness

This might actually be harder than, say, proving some undergraduate math theorems because reasoning about mutable state (especially when you want it to be performant) is tricky. I might guess that it could be easier to use a model checker such as TLA+ for that (although that can only verify algorithm descriptions and not implementations) because they seem to be more built with these things in my mind, but I lack enough experience with it to be certain.