Too early for an internship?

MMF

New Member
Forgive me if this is in the wrong section.

I'm a first year Applied Math PhD student and will be mainly taking courses this upcoming semester. In particular, I'm focusing on numerical analysis and differential equations, as well as some programming courses (mainly C++ from what I gather looking at the syllabus). However later on, probably around my 3rd or 4th semester, I'll be taking some mathematical finance courses as electives. I'm pursuing a PhD out of interest, but as a career I hope to land a position on the street some day. My question is this, is it too early to try to obtain an internship after the first year given I'll be taking mainly DE, NA, and programming courses?
 
Never too early to apply. Most others (as in,your competition) won't have or be pursuing a PhD, anyway.

I do find it a bit strange to be doing internships during a PhD, though (and you should've done them earlier), as 1. it is effectively a job and you don't try and get internships at TwoSigma for your annual holidays while you're working at Goldman, either. 2. a PhD sets you up for an academic career and there's basically no other benefit to it. If you wanted to work in the industry, the earlier you start, the better for your career. If you want to do a PhD out of pure interest in the subject you're studying, that's a good reason, but to be planning a move out of academia the first year of your PhD it just sounds like you're going to be regretting the lost time, money and career opportunities later.
 

Daniel Duffy

C++ author, trainer
An advantage of doing a Phd is analytical, independent thinking, original research in general. Especially hard maths like PDE, numerical and functional. And it's fun.

Pencil-and-paper mathematicians are in the minority in industry. So learn C++ and you can't go wrong IMO.

"PhD sets you up for an academic career and there's basically no other benefit to it"
Not true.
 
An advantage of doing a Phd is analytical, independent thinking, original research in general. Especially hard maths like PDE, numerical and functional. And it's fun.
Sure it can be fun, but at, say, 100k a year lost in opportunity cost (plus whatever lost in the lag it caused in the start to one's career), I can come up with all kinds of alternative fun to do. I did do a PhD (in physics), and I don't personally regret it (but then again, I did not imagine it as a stepping stone to industry, but when I started the PhD I was still planning to make a career in academia, so my circumstances were quite different from the OP), but I would never encourage that as a route into the industry. You will learn much more in terms of useful, applicable and marketable skills in 4 years in the industry than you do while doing a PhD: Indeed, PhDs are hired to the same entry level jobs as Master's, which is just testament to the fact that a PhD is effectively valued at 0 by most companies.

I find arguments such as "independent thinking" etc. difficult: Not only is something like that impossible to quantify, it is not clear to me that my many colleagues with PhDs perform any better at their jobs (a somewhat measurable, albeing arguably biased metric) than those without, and in fact I can off the top of my head name several examples from my workplace to support the fact (as I imagine anybody could). Furthermore, even if PhDs did quantifiably possess more "independent thinking", it's not at all obvious that this was the result of completing the degree, or if rather people with such abilities tend to self-direct towards getting the degree. Finally, I would even challenge you on a(n average) PhD really generating original research: I know many who have never published yet still are able graduate; There's also massive differences in how different professors run their research groups: Some will effectively give a clearly defined problem to solve and provide guidance on how to go about it, where others have nearly zero contribution and the student will need to come up with the problems themselves.

There is one tangible benefit to doing a PhD: It looks better on your CV than a Master's and so getting interviews for that first job is slightly easier. If you know what you want to do early enough, you can do your Master's in the subject or otherwise build your CV to attract those interviews rather than pay in years of extra work.

"PhD sets you up for an academic career and there's basically no other benefit to it"
Not true.
Oh, ok. Thank you for the informative, well presented argument that took the discussion forward.
 

Daniel Duffy

C++ author, trainer
Mathematicians have more (intellectual) fun than physicists, possibly. Physics is so rigid.

"When a student comes and asks, "Should I become a mathematician?" the answer should be no. If you have to ask, you shouldn't even ask."
Paul Halmos

"There is one tangible benefit to doing a PhD"
And you can tell mama and papa.

"100k a year lost in opportunity cost"
peanuts when spread over a 40-year career.
 
Last edited:
You've quite a lot good points here. However, I still believe PhD training can improve someone's capability. Check this out: Summaries for Quantitative Finance This is basically a Math PhD's study notes for his job in industry. From his notes, you can tell how he picked up statistics, probability(which are not his field in PhD), finance, programming etc. on his own. PhD training played a big role here. But you are right, industry may not as much appreciate this skill -- pick up things on your own-- as it claims. I did see quite a few PhDs from top universities had hard time in industry. I've not figured out what held them back. Maybe being able to tell a story as a sales person, bringing more business is more important? Maybe collaboration is more important than autonomy in industry? In terms of economy-efficiency, PhD won't be a good path though.
 
Last edited:

MMF

New Member
Never too early to apply. Most others (as in,your competition) won't have or be pursuing a PhD, anyway.

I do find it a bit strange to be doing internships during a PhD, though (and you should've done them earlier), as 1. it is effectively a job and you don't try and get internships at TwoSigma for your annual holidays while you're working at Goldman, either. 2. a PhD sets you up for an academic career and there's basically no other benefit to it. If you wanted to work in the industry, the earlier you start, the better for your career. If you want to do a PhD out of pure interest in the subject you're studying, that's a good reason, but to be planning a move out of academia the first year of your PhD it just sounds like you're going to be regretting the lost time, money and career opportunities later.
At the university I'm at it's quite common to get an internship it seems, lots of students either stay as an RA or go off to a lab/bank/tech firm for the summer. A good portion of the graduating students now work in industry as well.

An advantage of doing a Phd is analytical, independent thinking, original research in general. Especially hard maths like PDE, numerical and functional. And it's fun.

Pencil-and-paper mathematicians are in the minority in industry. So learn C++ and you can't go wrong IMO.

"PhD sets you up for an academic career and there's basically no other benefit to it"
Not true.
Exactly what I will be doing, lots of PDE, Numerical Analysis, and coding.

You've quite a lot good points here. However, I still believe PhD training can improve someone's capability. Check this out: Summaries for Quantitative Finance This is basically a Math PhD's study notes for his job in industry. From his notes, you can tell how he picked up statistics, probability(which are not his field in PhD), finance, programming etc. on his own. PhD training played a big role here. But you are right, industry may not as much appreciate this skill -- pick up things on your own-- as it claims. I did see quite a few PhDs from top universities had hard time in industry. I've not figured out what held them back. Maybe being able to tell a story as a sales person, bringing more business is more important? Maybe collaboration is more important than autonomy in industry? In terms of economy-efficiency, PhD won't be a good path though.
So if my main goal is to become a quant (preferably on the research side), admittedly for the big salary, would an MFE be a smarter path?
 

AppliedMath

New Member
So if my main goal is to become a quant (preferably on the research side), admittedly for the big salary, would an MFE be a smarter path?
No. Not to me. "Research" doesn't seem like a term to be taken lightly in those jobs. A quick look at 10 or so, while show they are a stretch for an MFE grad. Many, unfortunately, don't really consider MS-only applicant. They do, but you'd need a nice portfolio to show for it... like some decent research. Your MFE-thesis likely wouldn't be enough. If an industry research-based position is your goal, then my opinion is that a PhD would teach you that skill-set.

If you satisfy your curiosity, get the great research skill-set, and land a research-based role in industry before finishing the PhD, (e.g. by doing internships ASAP as you planned) then you could quit the PhD program early. It sounds like it's not really about the degree anyway - just the learning environment and direction it provides.
 
Top