[mga226]

An alternative view on this, which may seem cynical but which I find empowering:

In their best form, these types of questions are a litmus test for whether you have a certain type of background. When they fail it's because they test for skills not required by the job. It's fashionable to complain about this, but it seems to me to be more practical and useful to just get better at answering these sorts of questions. Leveling up from "terrible" to "adequate" at this will take less time than you think, and due to the craigslist penis effect[1] will put you way ahead of the game.

Easy resource to get started: https://www.interviewcake.com/

This book is pretty well regarded: http://www.amazon.com/Cracking-Coding-Interview-Programming-...

Btw, anyone with better starting places for this, please jump in.

[1] http://www.iwillteachyoutoberich.com/blog/the-craigslist-pen...

EDIT: formatting

[helpless]

My personal view is cracking the coding interview is not a good quality book. I would prefer to read jon bentely book instead. or formally eva tardos or cormen algorithm book.

These companies has made computer science students simply dev ops who never use algorithm (except during interviews). So many computer science students don't go beyond cracking the coding interview. This is really sad.

[mga226]

Fair enough, but I feel I should clarify my comment.

I'm suggesting that if you feel the interview process is flawed even from the company's standpoint (i.e. not screening for the appropriate skills) your best approach might be to reverse engineer the interview (i.e. get better at answering the questions, instead of actually developing the [perhaps irrelevant] skills being screened for).

The Tardos book that I think you're referring to is a $100+, 850 page CS textbook with the goal of making you an expert on algorithm design. The book I suggested costs $30 at is focused on a specific goal of getting through the interview.

I'll grant, however, that there may be better resources for this (and I'd be very curious to hear what others think those are).

[lk145]

I agree that these interview questions are going to be around for the foreseeable future, so people need to just learn them. But it's still a waste of time -- you have a limited amount of time to devote to learning new technology. You can either spend your time building a cool side project with interesting languages, frameworks, or technologies that are highly relevant to your work or you can use it to study algorithms you will never implement outside of an interview.

When faced with this dilemma in the past I have picked the interview prep because it's necessary, but I'd rather have been building cool stuff that actually helps me do my job better.