I've done Telephony, Games and I'm currently working in a high performance context. 99% of the time, you don't need to be thinking about Big O

I think part of it is going through all those sorting methods to show quick sort is the best so that a) students who run into new sorts are better equipped to determine it most likely isn't better than quick sort and b) to show the process used to determine quick sort is better than the rest in case you have some other algorithm options you want to pick the best of.

And yeah, depending on what you do, some tasks never involve any of that, or when they do, they get offloaded onto a library or something that gives the solution to that step directly.

But, for example, if you have a collection, the question of array vs linked list vs tree is still relevant, even if you're just choosing a provided construct that is built on top of one of those. Each has their strengths and weaknesses depending on how the data is added, removed, and accessed.

And with how slow things are these days despite how much better the hardware is, I think there's a lot of successful software engineers and programmers who should be using that stuff from school more than they are.