What are your worst interviews you’ve done? I’m currently going through them myself and want to hear what others are like. Dijkstras algorithm on the whiteboard? Binary Search? My personal favorite “I don’t see anything wrong with your architecture, but I’m not a fan of X language/framework so I have to call that out”
Let me hear them!
(Non programmers too please jump in with your horrid interviews, I’m just very fed up with tech screens)
I think the interview I least enjoyed was with an unnamed big tech company.
It was the first interview of the day and the guy came in with “so me and my buddy have been trying to solve this algorithm problem for years. I’d like you to try and solve it for me.”
Like… Dude, that’s not a reasonable interview question! You should not use algorithm questions that you don’t know of any answer to in an interview. You’re effectively asking someone to give you a solution to something way too complicated of a problem without even a few hours to think about the problem or sit down with it on their own.
That sounds suspiciously like doing actual work for them.
I had one interview where they literally got me to fix their Sendmail server while I was there.
I seriously hope you got the job.
I did.
I was there 5 months before heading back to grad school.
I seriously hope you got the job.
Oh god I’ve had an open ended one like that only once, and you’re right it’s terrible. Those questions would be great things to tackle as a team of peers where you’re all working together without the pressure, but dude you hold our careers in your hands. Pull it together
Aren’t most questions like this are simply looking at what approach you try and not a solution? They’ve been at it for years so they can easily tell if you’re trying something that makes sense or something trivial even if they don’t have a solution or even if there isn’t one.
The problem is you’re effectively leaving “can I program and work through the kinds of tasks this job entails” and entering “how do you work through a complex theoretical research topic” land.
White board questions should be relative softballs related to the work you’re actually doing to see how you think… Now that’s often forgon for “welcome to a game of algorithm and data structure trivia!” but this is just a much more extreme version of that.
Also if you don’t actually know the answer, how can you judge the direction? Even if you do know the answer for a problem that complicated, can you say the interviewee isn’t solving the problem in a novel and possibly better way?
I presume he was looking for specific terms like DAWG (directed acyclic word graph) and things like that as well… Which I know because he would teach me the names of things as I slowly rediscovered them in conversation. Personally, I don’t put much stock in grading someone on their knowledge of obscure data structures and algorithms either.
When I give interviews, I’m more concerned with the process than the results for some questions. I don’t really do it any more, but I’d sometimes ask one question not related to programming or anything on their CV just to see how someone works through a situation given a little bit of a curveball.
Sorry, but the answer we were looking for is “I’ll need to work on this over the weekend.” That’ll be all for today. We’ll call you.
Sometimes you don’t have to solve it but instead ask them about what approaches they tried so far and suggest a new one. Just showing interest
That sounds suspiciously like doing actual work for them.