3 ms·

I think you make good points. I also think that Leetcode problems are not good for this (at least for most companies) simply because the problems are too hard a

I think you make good points. I also think that Leetcode problems are not good for this (at least for most companies) simply because the problems are too hard and require preparation or luck. Some of the algo's that are being asked took quite some time to initially develop, e.g. Maximum subarray problem. In my opinion it would be better to ask something a lot simpler, should still give you this information.

twawaaay 4y ago
In general this is a mistake of interviewer asking question they would not be able to answer themselves. It is super easy to get this wrong. A problem seems much easier when you know the solution. Also, as interviewer you don't want to make mistake of comparing the candidate's knowledge to your experiences -- the candidate might be completely fine but just happened to have different experience from yours and did not meet the same problems as you. I have a small set of problems which I honed over the years. Some of them are problems which I got when I applied, which makes it easier for me to understand how it is when you are on the other side. Every single problem I have solved in every way I can imagine so that when I interview the candidate I can focus on the other stuff that I care about. I have developed understanding of where the candidates get stuck and for what reasons and how to best provide hints so that I can keep the session productive. When the candidate is stuck for a long time I tend to view this as my failure (nothing is happening == I am not learning anything) unless the candidate is really bad. What I am trying to say is that preparing good problems for coding interview is a hard task and badly prepared problems are probably why so many leetcode interviews are frustrating and inefficient.