Folks,
As promised, here are the collected questions. In addition to asking the jobs-list folks, I also asked cs-faculty what they thought. I have removed duplicates (or near duplicates) from the list, and I've taken a quick pass at organizing them around themes.
Some questions really stand out. For one, you better have something of an answer for "What are you going to be doing N years from now," or you'll be saying "I don't know" a lot. :-)
As promised, I'm also sending this list out bcc'd to the faculty that contributed.
Thanks all for your help!
-brian noble (CMU grad)
What have you done? Why? What's cool about it? Which of these things were fun? Which of these things excites you? Do you expect to continue this work? How?
A question the interviewer asks him/herself while interviewing: Can this person coherently describe their dissertation work, especially if it's not an area I'm very familiar with? Did they sound like they picked a fundamentally interesting problem to work on? Did they research it intelligently? [A bit too late for most of this -- brian :-)]
What are the lasting contributions of your research? (Also phased as: "What is your contribution to the science of computer science?")
What are the broader applications/implications of your work. (who else could use this approach/ideas/results, if you can't dream up a constituency outside of your own narrow theme group you don't have much imagination. but if you can't ground it in a reasonable way, then you don't have much perspective)
What did you learn from your thesis research? (don't tell me about the thesis results)
What were the major considerations for performance evaluation, eg. what could you directly measure, what simplifying assumptions did you have to make. (don't tell me that it was a perfect empirical experiment)
Let's say your have developed a new technique that provides an N% speedup. I would ask: Given that the new technique is harder/more complex to implement, for what value of N would you have decided it wasn't worth hassle ?
Of course in the old days we had a tradition of nearly all the faculty attending all thesis defenses, even when they were wildly outside their area of expertise. Nico Habermann would predicably ask "How would/did you verify (the program correctness) of your thesis". This was a particularly difficult question for people in AI to answer, and usually resulted in some tap dancing in the front of the room. Of course Nico knew that answer was "I didn't" or "I don't know how to" but, of course, that wasn't the question.
What's your favorite part of building a system? What was your part of the project? (for systems people working on big projects)
What is the most important problem to be tackled in your area now? In ten years? Whose work do you think is best in your field? What do you think of X's work? [This one is particularly insidious -- brian]
How much do you know about X? (where X is an area with which you are vaguely familiar)
"How does your work relate to my research?" (for some random research in a totally different area from yours)
What do you think will be the most interesting developments in computer science (not just your thesis area) over the next five years?
What are the five coolest things that have happened in CS while you were in grad school?
If you had a 1000x improvment in computing power, memory capacity, and networking power (which you will, in your lifetime, almost certainly), what "hard" problems in computer science "just go away," and which ones don't get any easier?
A question the interviewer asks him/herself while interviewing: Can this person rapidly figure out what the issues in my project are? Are they able to generate insightful questions, or for that matter offer insights into possible solutions? That is, can they think on their feet?
What do you envision will be the most important issues in CS (or in your subfield) 10 years from now? Where do you see yourself in N years? What do you want to achieve in your career?
[ For the most part, one of the weakest answers is a recitation of the "future work" section of the thesis. The questioner is usually trying to see if you have a broad enough view of the field to make sensible choices about what are now the important problems in the area. Sometimes the most important problem is an extension of your thesis -- but this is rare enough that it needs to be backed up with a strong analysis. It's also a good idea to relate your future interests to the department where you're interviewing, but to the extent that you compromise your better judgment. --- a faculty member]
What is the biggest accomplishment of your life? What was the biggest failure of your life? What did you learn from it?
What words would your references use to describe you?
[ I think this question is far more interesting than "how would you describe yourself..." --- the respondant ]
What areas will you personally have to work on in order to do this job to the best of your ability?
What are you best at (management / research / software / hardware / working with people / teaching / ...)?
We like to hire people who are a little bit strange. What's strange about you?
Give an example of a conflict you experienced with another person and how you resolved it.
How (if at all) do you like to work with other people? Lone wolf, part of a team of peers, leader of a group? Do you like to implement your own ideas? What characteristics do you value in a boss? What qualities do you value most in your team co-workers?
What did you learn in graduate school?
Why do/don't you want to...
If you won a few million dollars the lottery and had no constraints, what research (if any) would you do -- after you get back from the customary six months at the {beach, mountains, cabin in Montana...}?
What courses do you want to teach?
Are you more interested in teaching or in research?
Is funding available for research in your area?
What are thesis topics you'd suggest to new CS Ph.D. students who are begining now?
Here at XXX the grilling question from hell is: "Which Programming Language would you use in CS education". If somebody gives the wrong answer he/she is immediately burnt at the stake and remains history.
I only remember one silly question from my job hunt. The meeting with an senior associate Dept. Head was also attended by pretty wierd AI faculty member, named XXX. He started a meeting which he labeled "about my broader" abilities with the following question. "In your CV I see 8 conference papers but no books and no journal papers whatsoever... how do you expect to get tenure with a record like that? Is your research not good enough for journals or books?"
How will you make your final choice from among your offers? What is your most important criteria for a job? What is your ideal job?
Where else are you interviewing?
Do you have any questions for us?
The one I spend some time fantasizing about is "What would it take to get you to work here"?