Can I pursue my pHD in computer science?
July 26, 2021 11:29 PM   Subscribe

I'm older (36) and received my BS in International Business in 2007. I've been a professional software developer since then and have been more and more interested in theory. I've also enjoyed teaching through a bootcamp. I'd like to pursue a doctorate in computer science, is this possible?

I don't have a really strong, formal mathematics background, I think I took several statistics classes (applied probability?) in college. However during college I made it through most of Art of Computer Programming by Donald Knuth on my own, I sort of fell into programming and felt that if I was going to make it a career of it I needed to understand it. I really understood ... maybe 50% of it, it is fairly dense. Stanford had their entire under grad CS curriculum online and for what it is worth I went through that, but still felt overwhelmed by some of the advanced topics.

In any case, for some reason I started recently getting into some advanced projects on the side like trying to write my own version of emscripten, etc. and understanding the theory behind things. I was always more interested in reading things like Martin Fowler's design patterns or Codd's paper on normalization than I was in writing the software.

I recently saw someone get a doctorate in a hard science at around my age without an undergrad in that discipline, so it is possible but how feasible is this? How do I go about pursuing this? I've been out of academia so long that this feels a bit of a reach.

Are there other avenues I could take if I'm so far not qualified for a pHD that would be a good fit?
posted by geoff. to Computers & Internet (7 answers total) 3 users marked this as a favorite
(Caveat: I'm a grad school dropout, not a prof!)

My main concern with your question is that I can't tell whether you know what research is, and whether you'd actually enjoy it. Also, CS is a fairly broad field, and you generally don't just apply for a PhD in "computer science". Instead, you'd apply for one intending to specialize in computer vision, or machine learning, or languages, or human computer interaction, or robotics, etc.

Do you have a sense yet of what you're interested in specializing in? Do you have questions that you're driven to find answers to? Is there an unsolved problem that you want to spend a few years whacking away at?

For your "other avenues", I'd like to encourage you to look at doing a batch at the Recurse Center. It is a place to work on becoming a dramatically better programmer, whatever that means to you, and your "advanced projects" fit right in to the type of things people get up to there. So, it might be a good place for you to to e.g. spend time engaging in some corner of the academic literature and working on toy implementations of whatever interests you. This might help you figure out whether you want to take a sharp turn towards academia, or whether you just want a different type of non-academic job.
posted by Metasyntactic at 12:03 AM on July 27, 2021 [13 favorites]

Not trying to dissuade you, but it sounds like you may not have realistic expectations of what a PhD program is like. If you start out in a master's program, you can scratch this itch without all the pathos of a PhD program, and you can get a taste of the research life that awaits you as a PhD student, and then decide.

Source: A guy who did eventually finish his CS PhD, but at huge opportunity cost and toll on mental health...
posted by qxntpqbbbqxl at 12:11 AM on July 27, 2021 [7 favorites]

What I have been told is that CS graduate programs care more about candidates' "mathematical maturity" than they do about their undergrad CS background. Your professional experience should offset that some, putting you into a category of serious applicants, but you may have to take some mathematics courses first. I think that, if you could get through Knuth, you should be able to handle the work academically, but you may need some recent, documented success in mathematics courses, to overcome this kind of worry on the part of admissions committees.
posted by thelonius at 12:14 AM on July 27, 2021 [2 favorites]

I second the recommendation of the Recurse Center as a semistructured activity and group setting that can help, as Metasyntactic says.

I sense from your question that perhaps what you want is structure, instruction, and a peer group to help you learn CS theory; you could achieve this through a doctoral program, or you could pursue it through RC, a master's or postbaccalaureate program, community college courses in CS, or a local or online study group plus a paid tutor.

Maybe I'm wrong about what you're seeking. Going to graduate school to pursue a doctorate in computer science has several components, such as coursework in CS theory, reading other scholars' research papers, coming up with research ideas, carrying them out, collaborating with other researchers, teaching, learning about how to apply for grant funding, finding an advisor and creating a good working relationship with them, writing a dissertation and orally defending it, etc., and then emerging with a particular credential. Try disaggregating the experiences and benefits of a doctorate; which of them, in particular, appeal to you? Do your goals line up with the goals of an academic program that's aiming to train and emit academics who do original research?

Dr. Lindsey Kuper's blog can be a good resource for learning more about applying to these programs, especially if you are in any way a nontraditional applicant.
posted by brainwane at 1:22 AM on July 27, 2021 [4 favorites]

You absolutely can do this. But I'd encourage you to set your sights on a master's degree, not PhD. It's a shorter time commitment (2 years, not 7) and there are many more programs out there including many aimed at folks with practical experience and/or are further along in their careers. I'd start by looking at state universities and/or schools with programs aimed at people who are also working full time.

But the way you phrase your question makes me think you have more of a prestige serious academic degree in mind. I hope you're prepared for how abstract that will be! The way to pursue that is to choose a school or even better, a specific professor. Then start doing your own research aligned with them and try to make a personal connection. As someone with years of practical experience you have a lot to offer. OTOH you're also a non-traditional student so you have to work a bit to communicate your goals.

The real question is do you need funding? If you can pay for it yourself it's a lot easier.
posted by Nelson at 9:26 AM on July 27, 2021 [1 favorite]

Do you love doing research? Do you love teaching? Are you geographically mobile? If you answered yes to all three then go for it. I have a friend who earned a CS PhD in his late forties and is now totally loving research and teaching. He and his partner moved from the midwest to the southeast for his job. There is apparently a shortage of CS PhDs who really want to teach. It's a pay cut for sure, but the intrinsic rewards can make up for that.
posted by mareli at 11:27 AM on July 27, 2021 [1 favorite]

Best answer: Coming from a different angle, if you are looking at programs. Take a deep look at the CVs of the research faculty. I suggest this for multiple reasons: 1) it gives you an idea of areas of research being pursued and serves as a jump point for further research in a wide field that composes CS; 2) you will quickly get a feel for whether the faculty publishes with their students and carries out their role as a faculty advisor. You do not want to enter what is basically a medieval form of servitude to a person for YEARS who is not doing interesting research or making sure their students build their own research, and 3) it helps you triangulate which faculty you want to work with and how to gear your approach to applying to the program. There is no point in applying to a program that is not doing research you are interested in or in a faculty that does not advance your career/research.
posted by jadepearl at 2:58 PM on July 28, 2021 [1 favorite]

« Older Restaurants in downtown Cleveland   |   B-ing responsible Newer »
This thread is closed to new comments.