For those of you that had to answer a lot of "whats the Big O of ..." questions to get a job, how often did you need to use Big O analysis to actually do the work?

If their asking it must mean they use it. And the understanding of big o notation is important to them.

never.
if i have to do proper big O analysis, it means that by instinctual estimate it's not fast enough.

if it's not fast enough by instinctual estimate, then i don't have to do big O because i know it's not fast enough

@Midnigh-shcode but what about when you design the first implementation?

My thought was when you were the first code, you need to think about how it should be implemented, which Algo out data structure to use? Do you every think about the Big O?

That seems to be the rationale they give for knowing all the Big Os...

All I tend to care about is when to use a Set, List, Map, etc. Can always look up the details as needed. Just make it so it can be swapped out easily... but no apparently System Design is secondary to knowing how to implement algorithms from scratch...

Never know why... Knowing implementations as a reference helps but rarely do I personally ever think about actual Big O. Other than when something is slow in which I just tell others, "this process is slow, some monkey doesn't know how to code. I need to refactor and reduce the Big O"