Microsoft Job Interview Questions

[riferguson]

Three things:

First, those are questions for a variety of different positions all mixed together. Product manager candidates get asked different questions than engineering candidates.

Second, yes, Microsoft tracks the effectiveness of interviewers based on the outcome of the interview loops and the performance of the people who get hired.

Third, all of the questions listed, no matter how trivial they seem, are dispositive. I've interviewed engineering candidates with PhD's from top tier universities who couldn't reverse a linked list when asked, or who couldn't even explain basic concepts in their putative focus areas. You have to ask the seemingly stupid stuff -- it's a continual surprise.

Full disclosure: no, I don't work there now. Yes, I used to, a very long time ago.

[timr]

"Third, all of the questions listed, no matter how trivial they seem, are dispositive. I've interviewed engineering candidates with PhD's from top tier universities who couldn't reverse a linked list when asked, or who couldn't even explain basic concepts in their putative focus areas. You have to ask the seemingly stupid stuff -- it's a continual surprise."

While there are certainly idiots with PhDs, if you've got a candidate with a PhD from a top-tier institution who "can't reverse a linked-list", it's most likely because (s)he is under-prepared at the art of technical interviewing. Idiocy is clearly not impossible, but the conclusion that the candidate is an idiot should be on the low-prior-probability event list.

One thing that drives me absolutely crazy about technical interviews today is that most interviewers have completely lost the bubble on what they're trying to accomplish. It's become a bizarre, nerdy form of Kabuki theater, wherein candidates are madly trying to cram their heads full of list- and string-algorithm esoterica, while hoping that they're not presented questions so unfamiliar that they can't derive the answer in under 30 minutes, at a whiteboard. If the interviewer doesn't take this into account (and few do), the interview becomes little more than a random, high-pass screen, wherein lots of smart people are eliminated from positions based on bad luck, and little else.

Joel (on Software) has it right, but it seems like few people are listening: you want to make sure the candidate is smart, and can get things done. That's it. The goal is not to see if they're the next human incarnation of Alan Turing, and it's certainly not to see if they can derive fiendishly difficult algorithms during a whiteboard lecture (call me crazy, but I'm reasonably sure that Turing didn't come up with his theories while talking continuously in front of a whiteboard, while some asperger-y geek sneered at him from across the room.)

When you're interviewing, you want to make sure that the candidate can write code, that they're reasonably smart, and (IMHO) that they're not an asshole. It seems to me that our industry has taken "coding puzzle" and turned it into "crappy intelligence test", while largely ignoring the "not an asshole" part of the equation -- the part that's usually the most important in real life.

[arethuza]

I don't know how it works in the US, but in the UK a PhD is usually awarded based purely on research - there is relatively little cutting edge research that I am aware of that hinges on manipulating linked lists. :-)

[Locke1689]

I don't know what the author wants from Microsoft. I don't know about the other questions, but the technical questions seem fine. What does he want them to know if not how to do a preorder traversal without recursion? Name some obscure feature of Microsoft SQL? That will get you people who know specific pieces of software well, not good programmers. A good programmer should be able to quickly develop complex algorithms, irrespective of the environment or language. Everything else is pretty secondary to that.

I can tell you that my hardest question from my interview was developing an algorithm to get the greatest increasing subsequence from a sequence of integers. Will all good programmers be able to solve this? No, probably not. Will all the people who solve this be good programmers? Probably more likely. Microsoft is trying to find the best of the best and asking tough questions will get you much farther than random knowledge questions (which can be picked up in an afternoon with Google and documentation anyway).

[eclark]

Having been through this I can give a little insight. All the questions are from the interviewer. There is no MS book of interview questions. Each person asks whatever they feel like at the time.

While they seem like pointless questions for some jobs, this list is a conglomeration of questions for the three types of positions (Dev, Test and PM).

The questions are made to be very difficult because the interviewee is supposed to struggle and have to verbalize what he/she is thinking. If the questions are less strange there is the chance that a candidate will have seen them before and know how to solve them off the top of their head. That's only useful for weeding out people who can't program at all. The verbal communication is just as important as actually solving the problem correctly. Microsoft has a pretty high confidence in it's mentoring system, it believes that teaching the actual code is much easier than teaching how to think systematically.

I was given questions that took me > 6 hours to fully complete when I went home and did them. There is no way the interviewers actually wanted all the code, just a discussion.

Some of the "Gotcha" questions are from earlier in Microsoft's hiring practices, so take these lists with a grain of salt. Though they are wicked fun to try to solve them all.

HR at MS is very good at their job. They keep track of everything about who talked to whom.

[ambition]

I've been asked some of those questions at other top-tier tech companies (that the author isn't criticizing).

Conducting good interviews is hard. Watching someone attack a problem or write some whiteboard code is a not-bad system. Pick a problem that's too small or familiar and you're not filtering enough. Pick problems that are too hard or long and nobody will succeed. It's a balancing act.

[arethuza]

I don't know about you, but writing code standing up at a whiteboard while others watch just doesn't feel very natural to me.

Why not give them a PC and a small development task and leave them alone for a bit and then talk over their solution once it is complete?

[msg]

My problem with the coding questions (other than the fact that everyone has seen them) is that they're too shallow. So the questions are binary. If you can't do them, interview over, sure. But if you can do them they don't say that much. They're almost short enough to count as trivia.

I prefer asking for a program that will take about an hour, but will also get a range of correct answers and mistakes. The ideal question has maximum dispersion, like a good hash function. You can observe a variety of strengths and weaknesses, and the horrible candidates will cry, the mediocre candidates will cringe, the good candidates will nod, and the strong candidates will laugh.

[ambition]

Could you give some concrete examples of the sorts of programs you mean?

[bmalicoat]

This is a pretty baseless post. Like riferguson said, the questions span many disciplines hence some are technical and other are very broad. The OP is wondering if Microsoft's practices work? No, no company is hiring 100% coding geniuses all the time. However, Microsoft, despite their public appearance sometimes, always has some of the smartest programmers and researcher alive today working for them.

I was interviewed there a few weeks ago and the questions I got were much more related to the team's tasks. Most were coding but a few where design architecture but all were quite relevant (ie no linked list stuff, though there was a BST question).

[RyanMcGreal]

>How would you explain what a database is to a 5-year-old?

>How would you explain computer networking to a kindergarten kid?

I have had to answer both of these questions - but to an actual five-year-old, not to recruiters.

[Zak]

Write code that returns the length of a string without using any built-in functions.

It took me a while to realize that they're probably not counting infix operators that behave like functions except for the syntax. Accomplishing this task in Lisp or Haskell would be nearly impossible.

[Locke1689]

Doesn't seem all that tough to me

  my_len_acc :: [a] -> Int -> Int
  my_len_acc [] so_far = so_far
  my_len_acc (_:xs) so_far = my_len_acc xs $ so_far + 1
   
  my_len :: [a] -> Int
  my_len some_str = my_len_acc some_str 0

[ambition]

That's disingenuous. When they say "no built-in functions", they really only mean "don't use a library function that makes this task trivial." Which eliminates 'strlen' in C, 'length' in Haskell, 'len' in Python, etc.

[msg]

  (defun strlen (str)
    (loop for i from 0
          for char across str
          finally (return i)))

Works in my emacs... I (require 'cl).

[scdlbx]

Then don't write it in Lisp or Haskell. If they don't specify a language, use one that makes sense.

[chris123]

Aren't these questions mostly a way to trim the prospect pool to a size that you can do "real" interviews with? ("Real interviews" are the interviews after the ones in which you get asked these kinds of questions).