I had a very bad experience interviewing with Yelp. I met with a recruiter at my university career fair and was soon after invited to take a 20-minute programming challenge online. The question was very easy; it involved reading in N words and then deciding if all words were anagrams of each other. The issue was, they said to read in N words, where N was not specified, and they did not specify how to know when the user would be done entering their words. So, in my code, I first asked the user to enter how many words they would be entering, N, and then to enter all of their words. My solution was complete and optimal, however, their autograder did not like my answer.

I told the interviewer that their question was not well-described because I had no way of knowing when the user was done entering their words without first specifying N. The recruiter said that he agreed the question was vague, and instead of looking at my complete and optimal code, they asked me to take yet another technical challenge. Really, I just answered your vague question perfectly!

Anyways, I agreed to take another challenge. The second challenge was to read in strings and essentially sort them. Super easy, yet again. But the problem remained: how should I specify when to stop cin reading in? There was nothing in the problem specifications whatsoever, so this time I asked the user to type 'Done' after typing all of their strings. The autograder, of course, did not like this, even though my code was again complete and optimal!

So the recruiter told me I did not move on to the next round. It is ridiculous that anyone with eyeballs could have looked at my code and seen that I had given them the exact right answers they wanted. I am very surprised that such a big company would be so bad at this process.

Read in N strings and determine if they are anagrams of each other.

Ex) 'cat', 'act', 'tac' -> true 'cat', 'bat', 'act' -> false