This code is very easy to read, but I don't think it qualifies as best practice. For every character in the string, 'cnt' is checked to be less than zero. However, that check only needs to be applied when the character is ')'. For large strings, this solution will not be as efficient.

I had not considering using lambda.

With 6 seconds of time for the thread, it may be an extra challenging to try a genetic algormithm. Perhaps, impossible....

Yes, I can imagine I misinterpreted the instructions, but I don't think I did. Others code warriors here think the last test is incorrect.

I agree that the last test case is faultly. If you look at the raw data and follow the requirements, 'c' is the winner instead of 'e'.