[rplst8]

What you should learn in college is how to teach yourself to always be learning. Or at least to be aware of the trap doors that await when you try to solve a problem in the real world.

Learning the Big O Notation performance of algorithms that traverse data structures isn't so you memorize all the different running times in perpetuity. It's so you realize that performance and the type of operations you want to commonly execute are something you need to consider for the data you have to work with.

If you didn't learn that from your college CS courses, I don't want you working on my team.

[malfist]

The other thing college shows is that someone is capable of committing to a long term goal and seeing it through. That's important.

Note: I am not saying non-college grads cannot commit to a long term goal, there's plenty of reasons people don't or can't go to college. I'm only saying that a college degree is a signal.

[turdprincess]

That was my point as well - its not what you learn in college, its that you can learn something. I am not a proponent of Leetcode (and suck at it personally), but I think the same concept applies - just shows you can learn an arbitrary puzzle framework tangentially related to the job you are interviewing for.

Its also my experience that as you get more senior and focused on a specific skill (I'm an iOS engineer), interviews tend to focus on that skill more than leetcode. I think there are a few stalwart holdouts (facebook + google), but I have interviewed with several trendy big tech companies and none of them asked me pure leetcode - it was always framed around my iOS skillset.

However, for a junior engineer there is no way to evaluate this kind of skillset since it doesn't exist - so Leetcode is a resonable substitute.