Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Will I be rejected as overqualified?

  1. Jul 18, 2010 #1
    Hi, I'm beginning my physics PhD program this fall because I want to do astrophysics. However, even though astro is an interesting field, I have no intention of staying in academia afterwards. I'd like to do software in game physics or animation physics in the industry. Will I be frequently rejected as overqualified? Those positions are usually open to CS, physics, ME majors with "advanced degree preferred".

    I wouldn't really mind leaving with an MS if it significantly improves my prospects, but it's always been one goal in my life to earn a PhD. Or, is there anything else I can do other than dropping out at some point? Like industrial internships? (Do physics professors disdain non-academic internships?) I already have 1.5 yrs solid experience doing game physics.
  2. jcsd
  3. Jul 19, 2010 #2


    User Avatar
    Science Advisor
    Homework Helper

    Generally a physics PhD (with experience and interest in programming) is the prefered background.
    CS PhD's are rather looked down on in the software world. The general view is that they are academic ivory tower types and if that if they were that smart they would have done maths/physics PhDs.
    This isn't necessarily fair, but i have worked at a few places (as a programer with an astrophysics PhD) where their hiring wishlist was :
    maths PhD
    physics PhD
    other science PhD
    maths BSc
    physics BSc
    CS BSc

    Assuming you also have a background and interest in programming - this is perfect.
    The other big consumer of programming heavy physics PhDs is finance, so you aren't going to starve.
  4. Jul 19, 2010 #3
    If you have no industry experience, then no you won't.

    Also being rejected as overqualified is not a bad thing. It usually means that the employer thinks that you can get a better job elsewhere, and it's good for you if they turn out to be right (which they usually tend to be).

    Unless the game industry is radically different from any other computer related industry that I know about, then the Ph.D. will be regarded as at worst neutral and at best a strong positive. One bit of warning is that game companies have a reputation for being particularly awful places to work.
  5. Jul 19, 2010 #4
    This is ridiculous.

    Perhaps specifically in the field of game physics, you might have a point. I have no experience in that particular field.

    But in general, how anyone can have the idea that having a Ph.D. in a field unrelated to the work is better than having a Ph.D. related to the work boggles my mind.

    (Yes, you can take a Ph.D. in physics and get a job as a programmer or a quant. But if these are the jobs you want, you are better off preparing for them properly and studying either CS or finance.)
  6. Jul 19, 2010 #5
    It's true.

    The trouble is that having a CS Ph.D. isn't particularly related to the work. There is pretty much no correlation between one's ability as a CS researcher and that as an application programmer. You can be an absolutely brilliant CS researcher and be totally unable to program "hello world."

    A good analogy is that you can know everything there is to know about Navier-Stokes equation and be a lousy plumber, and vice versa. Just because you have a Ph.D. in electrical engineering doesn't mean that you are a competent electrician and vice versa.

    No you aren't, and I have the advantage here that I've actually gone through resume and done interviews on this.
  7. Jul 19, 2010 #6

    I must admit two-fish, I've wondered why this is myself. I trust what you say and know that you've had first hand experience and that you can speak on this topic at length; I'm just not sure why companies would rather hire a physics PhD over say a CS PhD, especially when the work seems to be CS related.

    I do understand what you said about being a CS researcher and a good programmer are not interchangeable, but being a physics researcher and a good programmer are not interchangeable either...

    If you had to answer the question of why do employers like to hire X PhDs over Y PhDs when the workload is most definitely related to the Y profession, what would you say?
  8. Jul 19, 2010 #7


    User Avatar
    Science Advisor
    Homework Helper

    A CS PhD might never have written a substantial piece of software, and any programming experience they have is likely to be in an esoteric formal logic type language.
    A physics PhD, especially one in theoretical physics has written a lot of code, as have most experimental physics PhDs - who probably also have experience in hw.

    CS degrees and PhDs are not widely admired. many of them are pretty poor. It's a young field in terms of academic generations and it changes a lot, so it's not really got a well established curriculum.
    If you are hiring for 'smart and gets things done' you have to wonder at somebody who decided to do CS rather than maths/physics - were they just looking for a job, or were they not as smart as somebody who does maths?
    Then somebody who went on to spend 4-5years proving some obscure CS theorem falls slightly short on the 'gets things done' topic - compared to a physicist who built a large lab experiment.
  9. Jul 19, 2010 #8
    I suppose it depends on what you did your dissertation research on, no?

    I mean, I can imagine that there's a ton of CS Ph.D. research going on right now in artificial intelligence/machine learning. I see tons of job postings for these types on phds.org.
  10. Jul 19, 2010 #9
    If a Ph.D. has always been a goal, then do it! When you apply for jobs, you can tell them this, and my guess is that hiring people will be impressed by your goal-oriented personality.

    Internships are always a good idea, I think. But you might be hard pressed to convince your advisor that it's a good idea.

    Also, if you want to be a game programmer, why not ask game programmers how they got into the field? We're all just a bunch of academics and erstwhile phd students :) I'd wager that no one posting in this thread has ever developed a game for anything but their TI-89s.

    Join LinkedIn, and find a group which specializes in what you're interested in, like this one, and stalk the group members and participate in discussions. This will let you network, so that in 6 years you will have an in in the industry. It will also keep you current on the field, so that when you apply for jobs you can use big words and impress the hiring people. If you KNOW what you want to do, your best bet is to seek out people already in the field, find out what skill set they have, and then work to tailor your graduate education towards that skill set. Your advisor will probably not object to you learning C++ or F# or whatever if you can convince him that you need it for your research.
  11. Jul 19, 2010 #10


    User Avatar
    Science Advisor
    Homework Helper

    A very specific research topic - eg a PhD in creating 3D characters from 2D video, with a working implementation in an industry wide language/platform - then yes thats valuable.

    A new approach to the Pootzle theorem with at most a 5 line program written in LISP - then no. I would rather have an astronomy N-body modeler that took a huge Fortran code from his supervisor and made it run in OpenCL on a GPU.
  12. Jul 19, 2010 #11
    I guess everyone's experience differs. I've never encountered this attitude myself in 15 years of working in Silicon Valley.

    If I had a Ph.D. in astrophysics and was looking for a programming job, I think I'd worry more about being considered unqualified than overqualified.
  13. Jul 21, 2010 #12
    Thanks. I'll try to squeeze in a CS minor (surely that will help?) and convince my advisor it's useful. I went and signed up for Linkedin. One thing that worries me is that I'm a foreign national so I'll need a visa such as H1B, then someone will need to explain why a PhD is to be hired for a position requiring only an MS. Any idea where I can find info on this?

    In my work experience, they don't really mind if a programmer's degree is in CS. Actually, only one programmer on our team is CS MS, while others are ME MS and I'm Physics MS. But it's just one company in this one part of the world... I will surely take some more CS courses, lest they think I'm unqualified based on courses taken.
  14. Jul 21, 2010 #13
    Nice to hear from an astrophysicist programmer! May I ask you what industry you have worked in? I mean, what kind of software did you work on and is it related to physics in any way? I have investigated the finance route, including reading through "my life as a quant" and quant interview guides, but honestly, Pixar is still many times more attractive than Wall Street to me!
  15. Jul 21, 2010 #14
    That's probably a waste of time, in my opinion.

    Just pick up a book. To you, a computer is a tool, not something to be studied. You don't care about the internal architecture. You don't care about writing your own compilers or assembly language or any of that crap. (Maybe you do, but it won't help you do physics. I highly doubt that people who program games use that stuff, but you should find out.) Look at the languages people are using in the field that you're interested in, and learn those languages. This may be different from what people are using in the specific group you work in (lots of physicists use fortran 90), so it might take more work. Most likely, you'll need to learn C++, and C++ will be around for a long time, I think.

    If you're dealing with large data sets (which you likely will be), and you have to do scripting, it's advantageous (to you) to learn how to do it in as many different scripting languages as possible. In those cases, your advisor won't care that you use Perl of oCaml or whatever.

    And keep current. You'll notice the trends in job listings. You should always keep your eye on job listings to see what employers want.

    These are all things that I wish I had known 4 years ago, when I was convinced that I'd be a faculty at some university at some point in my future :)

    Regarding the visa: most people don't care, especially in tech jobs. But if you can get some improved status (i.e. citizen or a green card), then you should shoot for it---it will only help your prospects, and it won't hurt (i.e. positive expectation).
  16. Jul 21, 2010 #15
    There is an obvious selection effect here. I don't get hired for jobs that I'm considered unqualified for.

    Since I have a Ph.D. in astrophysics, it's much more likely that I'm going to be employed by a company that thinks this is a good thing, then one that doesn't.
  17. Jul 21, 2010 #16
    Oil industry, logistics, wall street.

    One common factor is that I didn't have to explain the usefulness of the Ph.D. in industry because in all of the situations the hiring managers had Ph.D.'s themselves. All of the companies that I've worked for have had Ph.D.'s (although not necessarily in physics) in mid-management.
  18. Jul 21, 2010 #17


    User Avatar
    Science Advisor
    Homework Helper

    I was an experimental physicist, worked on big astronomical instruments so the switch to building big instruments in industry wasn't much different.
    Fulltime pure software jobs I have done massively parallel data-mining, LIDAR and 3D scan modeling for mining and oil, and now 3D video.

    Knowing the physics was useful for some 3d maths stuff (understanding matrices etc) and general numerical techniques (SVD etc).
    In general though the science degree was useful for the experimental nature of a lot of software design. Designing an experiment and designing tests are very similar.

    Computer graphics is one area where there are a lot of directly applicable CS PhDs - it is probably harder to get into than Wall St.
    One good area crying out for physicists that can program is industrial simulators, everything from machine driving to surgery simulators. These generally use games engines for graphics but need to model real physics - if you are learning to drive a 300ton haul truck it has to behave differently when it's full.

    I know one company that goes to the lengths of CFD modelling the oil flow in the transmission so the simulator behaves differently as it warms up and with the time since the last virtual oil change !
  19. Jul 21, 2010 #18
    Point taken. And since I have a Ph.D. in CS, it's much more likely that I'm going to be employed by a company that thinks this is a good thing than one that doesn't. :smile"

    And while we're on the subject of selection effects...
    It's been exactly the same with me.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook