More Valuable To Programming Employers? Math PhD Or XP?

[someperson05]

I'm finding myself in a precarious career position. I'm working on a PhD in New Zealand in a very particular and odd foundations of mathematics theory. This theory is not widely regarded as being of much use and I was hoping to discover that it could actually be very useful. However, after a year that prospect is becoming less and less likely as I continually poke holes in multiple attempts to make it work. My career plans after graduation was never to do anything with the PhD, I was in it for the research and I was leaning quite heavily into doing programming as a career. I don't think that this research itself will help me in any programming career other than perhaps having a fluidity with alternative logical concepts. (For those familiar, think constructive mathematics but even stranger and not useful in computer science.)

For background, I've been programming since I was about 12 and have some resume bullet points in it, including 2 summers of internships in image processing during my college years and helped start a game design course in C++ at my high school (and effectively helped teach it). Picking up new programming languages is now easy for me. I also don't have a Masters in math, only a Bachelor from the US. I managed to get a scholarship without it.

Right now I'm weighing my options and trying to figure out what would most likely offer the best long-term gains. Here is my current plan: Stay with the PhD for another 6 months to make sure that I don't make this decision on a fluke. Really see if nothing will come out of the research. Then here are my options:

Option 1: Try to get into a coding bootcamp or find beginning programming jobs through contacts who are well into the programming field.

Option 2: Finish my PhD and then do option 1. This basically puts me back about 1.5 years. Thus, it essentially amounts to 1-1.5 years extra experience.

Option 2 looks unattractive to me for many reasons. The PhD thesis would be composed of why all the stuff I've been working on for the first year doesn't go anywhere. I don't know how psychologically healthy that'll be at the end of the 1.5 years of more research and how much willpower I'll actually have left over to see myself pushed again. This research has been particularly stressful as single results can essentially make the rest of it uninteresting.

What is the essential economic benefit of having a PhD in (really really) Pure Mathematics vs an extra year of experience? Does it harm me in applying for jobs to say that I didn't complete a PhD for the reasons I've detailed (that the research flopped)?

Thank you very much for any assistance you can provide.

 

[Almeisan]

The PhD is about showing you have to ability to solve an academic problem, to do fundamental (or sometimes more applied) research.

Yes, it is possible that your work during PhD can be rolled out into startups, or be some hot new technology, or you develop abilities and skills the private sector is anxious to hire. But don't count on this.

Sometimes after four years of working on method A, the only thing you have discovered is that method A does not work and that people should discard it and not waste their time again (like you kinda did?) on method A, but look for method B.
You will have spend four years and all you have shown is that this is a dead end. That is very valuable contribution to science. But you do not have any skills or techniques that are ready to move into the private sector.

This is often ideal; research that reaches a point where new technology becomes profitable and the academic world hands it over to the private sector. This is often a great moment to leave academics and enter business. Doubtful it will happen during your PhD thesis.

The problem is, often 100 things seem uninteresting, mundane, or dead ends, but then one of them turns out to be a break-through. You never know which ones. Then, when there is a break-through, everyone jumps on that and people often just get the crumbs left over by the original people or by the top teams.

This is why a PhD is about research and about showing you can do research. You are showing that they can safely hire you to do research.
People that hire postdocs know that many fine researchers by 'bad luck' worked on problems that in the end didn't deliver. They care about how diligent your research was. Professors can be judged about picking the wrong research subjects, PhD students can't.

You won't get to lead a research team in private industry ever without a PhD(or without nice publications). PhD is what allows you to get research jobs that allow you to get better publications. This is why people go on to get PhD's as a career decision (besides the intellectual reasons).

I am sure there is pure math research being done by some companies, but that's probably rare. I am sure Google wants people with PhD's in math, not an MSc in CS, to work on certain problems. I can imagine these guys would do some 'proof of concept' or consultancy, then hand it over to coders to create the actual product. And it's probably just a few guys(or girls) at google HQ.

A programming job? You don't even need an MSc for that. You just show your portfolio. Heck, even if you have no degree but somehow a great programmer, which is easy to show through portfolio, can get you jobs.Would it hurt to inform people you quit your PhD 'because your research flopped'? Maybe if they know something about academics, they would find that a bit of a silly thing to say.

To me a math PhD and a programming job are just so completely different, both in what stuff you do and at what level you do stuff. I assume you want to do coding if you want a job from a programming because I highly doubt an ordinary software company has much use for math PhD people.
But that might depend on your expertise. I mean, by now you kinda know what applications the stuff has you worked on, right?
I mean, you can probably code but you are a mathematician at heart. So at least get an idea where your MSc-level knowledge might be applied. Surely, most CS people and many really good software engineers are not very good at real math.
If all you want right now is to get a software engineer job, a math PhD will be useless for that. Will companies hire math graduates that have undeveloped programming skills over CS engineers? Sometimes, but those math PhD's will have to be a lot smarter.
I think this myth kinda comes from the US where in the past it seems CS BSc degrees may have been considered weak by employees. If you want to hire really smart people, you hire those. Apparently some physics and math PhD students are really really smart, I don't know. If I were a business in either CS or finance, I would recruit among CS and quantative finance/econometrics PhD graduates, if PhD level skillset is what I need. If I need a programmer, I get one of those. If I need a mathematical problem solved, then I get a mathematician.

You should be able to get out of a PhD program halfway and get an MSc. At least, that is how it usually is.
Usually one completes a MSc and get a job as a PhD candidate/research assistant. Or you get a BSc and become a grad student working towards a PhD. Two very different tracks. Usually, you wouldn't be able to get into the former as an BSc student because you need 1.5 to 2 years more of classes before you can actually start to work on a problem and start your thesis research.

As for your research right now. Don't you have a supervisor? What does she/he think? If this is a dead end, why not finish this dead end, publish the result and move on to something more promising (which might also be a dead end). I don't know how common it is for a PhD to be a collection of dead ends.
I sometimes hear stories from the US where people work 7 year on their PhD, suggesting to me they keep researching until they find something 'nice'.
But where I live you just do your job well for four years, then your contract(and pay) ends, you clean up the loose ends and write up and defend your thesis.
Here they can't let you do four years of underpaid work, be very happy by all the nice research you did, and let you leave with nothing.
If they aren't happy with the quality of the research you have ben doing so far, they should have let you know. Usually, there is a point of evaluation and sometimes the PhD is changed or terminated at that point.

If you can't handle the psychological mind-**** of having to work on things for years only to find out it is a dead end, then surely research isn't for you. A math PhD wouldn't be the most appliable in the private sector. Doubt many companies are working on the Riemann hypothesis, ready to file a patents to make millions. What that means for you, no idea. You must have had reasons for going towards a math PhD all these years.

That being said, besides work at the math departments themselves, many scientists can really use a math or statistics collaborator. Hard to know the ins and outs of math techniques, you are never going to know you are going to use once you need them, if your own subject is already broad, deep and diverse enough. Those people won't care if your research 'flopped' because you did all things right, worked dilligently and discovered the underlying thruth.

As for economic benefits, I doubt you will be making less in 10 years because you have a PhD in math, once you get the right job for you. The problem is, the more highly educated you are, the harder it is to find a suitable job. Often, the job market includes the whole globe.
If you want to go for that niche inside the private sector that employs MSc and PhD mathematicians, go for it. Once you make it there, you won't have a poor salary for sure. But job security may be a problem, as well as moving around a lot.

Will you be making more money trying for good PhD math jobs than someone with just a BSc in an applied field, that moves up to management fast? Probably not.
Will you be making more money than someone with a BSc whose career kind of duds and who competes against MSc( and PhD people) with better job experience in the same field? For sure.
Will you be making a lot of money with your PhD in math when you stay in your own home town, working jobs outside your field (like programming)? Probably not.

What are your options about finishing current dead end, publishing it, following a few more courses in CS and graduating with an MSc in CS/software engineering instead of math?
It is actually a good thing for an employer to hear that you won't jump ship at the first oppertunity of a research job in math at PhD level when you apply for CS jobs with a PhD in math, because that is what they fear; you are overqualified and you'd rather be doing what you were trained to do.

 

[jkl71]

Based on my experience a PhD in pure mathematics won’t help you at all, the extra year of experience means you’ll be in intermediate/advanced positions a year earlier than you would be otherwise. On the other hand, if you have (say) a 30 year programming career, how much of a difference will one less year really make?

My background is I’m in the USA (so, everything I say is based comes from that perspective) and have a PhD in theoretical physics (in terms of practical usefulness it seems pretty similar to your Math PhD work). I spent a little over a year as an untenured lecturer then went into software development. My research didn’t involve programming at all, so you’re way way better prepared for a career in software than I was.

Does it harm me in applying for jobs to say that I didn't complete a PhD for the reasons I've detailed (that the research flopped)?

I don’t think not completing your PhD will harm you, but I wouldn’t tell interviewers it’s because your research flopped. I’d phrase it that you didn’t like academic work that much and wanted to write code. This actually sounds like an honest description from how I read your post. In general while interviewing I’d stay away from negative comments and focus on the positive.

I actually think not completing the PhD may help with interviews early in your software career. When I was first starting software, interviewing was tough, I faced a lot of skepticism. One attitude (like what Almeisan mentioned) I encountered was that that I was only doing software as a temporary thing and that I’d leave as soon as a cool physics job came up (several people that interviewed me seemed to think this was likely to happen, they just didn’t understand the market). Another was that I didn’t really want to work in software, I was just doing it for the money (IMO, this isn’t bad, but some people think it is).

If you tell interviewers that left grad school because you want to write code, you’d bypass these questions before they came up and probably be seen in a positive light. I had one (and only one (he was a jerk)) interviewer suggest that I was a “quitter” for leaving physics, I guess that somebody might say the same to you, but I think most will respect that you can be pragmatic and be happy that you really want to be a programmer.

I also don’t think having a PhD will help you in the long term (might be different if it was an area of math with more practical applications). I’ve worked in research situations where people liked to have PhD’s on staff, but that was rare and even then it didn’t make any real difference. Since my first couple of jobs very few of the people I work with even know I have a PhD. In fact it’s such a small part of my resume, even some people that have read my resume and interviewed me didn’t notice. If I thought having a PhD was a plus, I’d do more to highlight it on my resume. For all practical purposes having a PhD hasn’t helped me at all in software, while the opportunity costs were pretty high.

In your situation the only thing I’d worry about is 10 years from now, for some reason, thinking, “I really wish I’d have finished my PhD”. That’s my 2 cents for what it’s worth.

 

[someperson05]

Thank you both very much for your thorough and long responses. They helped me really help to narrow down what was important in the choice I am making. As it looks right now, I think I'm going to finish out the PhD. In the scope of it, it isn't very much longer. I'm not sure if it is the wisest decision economic wise due to some of the skepticism that jkl71 mentioned, and my lower willingness to move around which Almesian mentioned. (I'm from the States, studying in NZ. Don't remember in if I mentioned that. However, this has made me realize I want to be close to family.) However, giving this issue some honest thought has reminded me that I'm doing this to try and get something out there. Some sort of contribution, even if it turns out to be of little value. I guess someone has to do it.

Thanks again.