Recently gave an SDE 2 interview where the problem was similar to Design Hit Counter.

I started with a brute force approach and then moved to the optimized solution. The discussion went well overall and I was able to explain the tradeoffs and reasoning clearly.

I implemented the solution too, but one edge case was failing near the end and time ran out before I could fully debug it.

One mistake I realized later was that I mentioned the complexity for get() as O(N) instead of amortized O(1).

For people who’ve interviewed at Uber or similar companies — how are rounds like this usually evaluated when the approach is mostly correct but implementation/debugging is incomplete?