17

I have in the next few days an interview for a post doc with a top UK university. The interview is on Skype, directly with the PI, in applied computer science.

I am preparing for that interview and I have a question. In your experience, how common are technical topic-related questions, like the description of a well-known family of methods, or of certain papers?

  • 13
    I suspect this is one of those (many!) academic questions in which variability in field / location / institutional culture makes a uniform answer useless or worse. – Pete L. Clark Dec 18 '17 at 22:23
  • 1
    I don't know why this is flagged as too specific: I found it quite general enough in answering. – jakebeal Dec 19 '17 at 20:10
  • @arboviral Right. I misread. – Leon Meier Dec 20 '17 at 9:23
  • Be prepared to be asked how to apply the Third Gödel's Theorem, whatever it might be. It's applied computer science, after all. Any doctoral student should know it. As an alternative, be prepared to be asked how to apply X, for whatever X from the book of Rogers "Theory of Recursive Functions...". All these questions are equivalent and fully, completely interchangeable. Or, if you want to an industry-near question, be prepared to be asked to name some common statement of C and C++ that evaluates to different values dependent on whether you take a C or C++ compiler. – Leon Meier Dec 20 '17 at 9:24
  • I don't understand the "off-topic" designation here. It's clearly on-topic; if the answer is "this varies a lot by field and culture", then that is part of the answer, not a mark against the question. – 6005 Apr 8 at 12:24
24

When hiring a person for a technical position (including a post-doc), any employer will want to have some confidence in the degree of technical proficiency of the applicant.

If the PI already knows you (e.g., through interactions at conferences), then they might not feel a need to ask technical questions. Likewise, if the PI has a strong relationship with your advisor and is willing to take a recommendation on trust, they might not ask technical questions. In pretty much every other case, you should expect some degree of probing to determine your technical competence.

What form that will take, however, is completely unpredictable. Evaluating technical skill in a short interview is a notoriously difficult problem, and people have lots of different idiosyncratic approaches to the issue. As a result, I would advise the following in terms of preparation:

  • Do not treat this like cramming for an exam. Do not do any study or review of core technical materials. Either you have the skills and competence that they are looking for or you do not.
  • Make sure you are familiar with the work of the PI and their lab, and think carefully about how your skills and background fit with their current directions (though you might get surprised by the PI moving in a new direction).
  • Make sure you are prepared to talk about your own work, your skills and background, and what you want to get out of this postdoc.
  • Get a good night's sleep and do whatever else you need to do to relax and be in a good and confident place personally.

Finally, remember that the goal of this interview is not to get yourself a job, but rather for both you and the PI to figure out whether you and the job are a good fit for one another. Make sure that in the discussion you are also prepare to look for the things that will help you be satisfied with this position as well.

Good luck!

| improve this answer | |
  • 9
    +1, except for, "Do not do any study or review of core technical materials." The people interviewing you will assume that you prepared for the interview, including studying. And there are always questions about the foundation of your field that are easy to ask but for which you might not know the answer offhand because your particular work requires you to focus on other information, meaning you no longer remember the specifics from your undergraduate classes. A few hours reading up on material you think you might be asked about is completely reasonable. – Kevin Dec 19 '17 at 5:26
  • 2
    @Kevin At a Ph.D. level, I would expect somebody to be generally conversant with all of the core material of their field. If they aren't, cramming for an interview like it's an exam isn't going to help. – jakebeal Dec 19 '17 at 20:11
9

One data point: my interview for my current position. Canada, not UK tho.

I was interviewed in a skype conference call with 5 professors, members of the project. They took turns asking me stuff, from "why Toronto" to "how would you solve missing data points or noise in the data". The whole process took about one hour, so the answers were rather broad.

One question that I believe you will be asked is something like: "Summarise your previous experience and how that relates to this position". I did a few other interviews, and that is a fairly standard question.

| improve this answer | |
1

It is very very unlikely that you need to describe (memorize) "well-known family of methods", but if you are supposed to know about them, you should be able to

  • discuss their properties, strengths and weaknesses;
  • compare them to an alternative family of methods;
  • how they may be useful in the PI's project
  • what are the potential challenges to implement them for the project etc
  • ...
| improve this answer | |
1

I have had technical questions in a postdoc interview in the UK, but it was for an interdiscplinary project and the questions were on the aspects I hadn't studied for years -- clearly I needed testing on those. They weren't inherently tricky questions but designed to check the areas I hadn't demonstrated by other means. I had a decent understanding and some knowledge (not just from reading their papers) but it wasn't enough for what they needed. The interviewers did it very well, as a pleasant discussion.

Postdoc interviews in the UK are very variable. Some institutions regulate them quite closely while others leave it up to the PI's discretion. For similar positions one group will ask for a presentation (given in front of the other candidates, with questions which may turn technical) while others will just have a panel interview.

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.