Suggestions for what to do with my Ph.D.

[grapesmoker]

This is my first post on PF, but I've been lurking for a while, and I hope the collective wisdom of the community can help me out in a moment of quarter-life crisis.

I'm in the terminal stages of finishing up my Ph.D. at a respectable but not superlative institution. I've been working in astrophysics, specifically in the area of instrumentation (even more specifically, I've built attitude sensors for a balloon-based telescope mission). I've determined that I don't want to remain in this field as a post-doc or try for a professorship or anything like that; I'm basically bored with the work and I'd like to do something else.

I want to be up-front about a certain less-than-desirable aspect of my CV, which has caused me a lot of personal grief. That is this: I don't have any publications. The major reason for this is that balloon-based missions have a very long life-cycle, and in order to collect enough data to actually do some science with, I'd have to be in grad school for another 3 or more years. I've already been here for 5 and I don't think I have 3 more years left in me. As I've said above, my work has been entirely in instrumentation (i.e. hardware, mechanical design, programming, etc.).

I'm considering a number of courses of action and I'd like to hear any input from others who might have gone through similar situations. Here are the general options that I'm thinking about:

1) Industry work: this is obviously the most alluring in financial terms (especially after the relative poverty of grad school). My question about this is: what kinds of opportunities are available for people with a physics Ph.D. I feel rather inadequate when I compare myself to people with, say, an engineering background, since where they focused on developing specific design skills, my work has required me to basically be a jack-of-all-trades to some extent. I can do a lot of things, and I even have a pretty good background in programming and certain design packages, but I'm worried about lacking whatever specific skill is considered hot right now. How have other people managed a transition like this?

2) Another branch of science: thanks to my friends in another department, I've kind of gotten attracted to computational biology and the idea of using my physics background in that context. Even prior to starting grad school, I had really wanted to do computational neuroscience, but for financial reasons I was not able to do this at my institution. Of course, I don't have a specific background in this field! I'm also worried that my lack of publications will kill me in any attempts to find work in another field of physics. Again, I would appreciate the insight of anyone who might have gone through this, and any advice on how to make this work.

3) Law school: this is another possibility that I've been thinking about. I know there are opportunities out there for people with mixed science/JD backgrounds. I'm worried about the fact that, while my undergrad GPA was respectable for the institution and majors that I had, it wasn't really all that awesome as far as law schools are concerned. My LSAT scores are solid. What I suspect this translates to is that the top schools are a reach for me, but I have a good chance at schools outside the top 14. What I'd like to know is whether having a science Ph.D. is any benefit in the admissions process, and whether not graduating from a top school would significantly impact my opportunities afterwards. Law school is a huge investment, so I want to be sure that if I'm going to do this, I know what I'm getting into.

4) Finance: I've recently heard the seductive call of various financial organizations recruiting on campus. I'm very attracted to the idea of mathematically modeling financial interactions but don't really know how to get started in this. If I decide to try this, how can I make myself an attractive applicant for quant positions and such?

5) Other: science writer, science and public policy, science and philosophy, ?: what can I say, I have a lot of diverse interests. I'm seriously open to any reasonable suggestions about what I can do with my degree.

I realize that this is a complex and fairly open-ended question; I'm basically trying to get a feel for some of the things that people pursuing various non-academic careers with a physics Ph.D. have done. Any and all information, suggestions, criticism, and so on would be super-terrific and much appreciated.

 

[colonelcrayon]

The law bubble has burst. I would be really careful taking on big law school debt.

 

[grapesmoker]

The law bubble has burst. I would be really careful taking on big law school debt.

Right now my plan is to only do that if I can get into a top program. Like I said, a reach, but if it happens it might make sense.

 

[mal4mac]

You may want to go to dinner parties, or on dates, and admitting you are a banker would not go down well in the current climate. Given the recent total failure of financial models, and the financial destruction of many not-too-well-off people, do you really want to dance with that devil?

 

[twofish-quant]

You may want to go to dinner parties, or on dates, and admitting you are a banker would not go down well in the current climate.

Sometimes you have to just not care what other people think. Personally, I look at myself in the mirror, ask myself some deep questions, and I've come to the conclusion that I'm actually doing good in the world, and as long as this is true, it doesn't matter much what other people think. After all, being a physics Ph.D. isn't the most attractive thing to bring up at dinner parties.

Given the recent total failure of financial models, and the financial destruction of many not-too-well-off people, do you really want to dance with that devil?

Some models in some places failed. Some models in some places didn't. One of the interesting things is that the models that people did in places that where physics Ph.D.'s were hired and encouraged to actually ask tough questions tended to work a lot better. A lot depends on the people you work with, and when went out job hunting a few years back, it was really obvious during the interviews which places cared just about the money, and which firms had people that actually cared about something else.

 

[twofish-quant]

4) Finance: I've recently heard the seductive call of various financial organizations recruiting on campus. I'm very attracted to the idea of mathematically modeling financial interactions but don't really know how to get started in this. If I decide to try this, how can I make myself an attractive applicant for quant positions and such?

Learn some very, very basic C++ and finance. Don't try to make yourself an expert in this, but just pick up a few books and start getting the basic terminology. If you have no C++ experience, then download Visual Studio C++ Express from the microsoft website, and compile "Hello world". Also do some math puzzles.

Also review your Ph.D. work, since about 75% of the interviews will be on that.

The other thing to do is to go to www.dice.com[/url] and [url]www.efinancialcareers.com[/URL] and start contacting headhunters. One thing about financial hiring is that people are *seriously* overworked, and everyone is just anxious to get new hires. There's also a ton of work that has to be done. Just to name one thing, there is a massive effort everywhere trying to figure out how to put credit and liquidity risk into interest rate curves.

 

[comp_math]

You mentioned learning some basic C++ below. This is in contrast to what I read on places like www.quantnet.com, where I read that people need to be skilled in things like templates, the boost library, etc. Could you comment on this? Thanks.

Learn some very, very basic C++ and finance. Don't try to make yourself an expert in this, but just pick up a few books and start getting the basic terminology. If you have no C++ experience, then download Visual Studio C++ Express from the microsoft website, and compile "Hello world". Also do some math puzzles.

Also review your Ph.D. work, since about 75% of the interviews will be on that.

The other thing to do is to go to www.dice.com[/url] and [url]www.efinancialcareers.com[/URL] and start contacting headhunters. One thing about financial hiring is that people are *seriously* overworked, and everyone is just anxious to get new hires. There's also a ton of work that has to be done. Just to name one thing, there is a massive effort everywhere trying to figure out how to put credit and liquidity risk into interest rate curves.[/QUOTE]

 

[twofish-quant]

You mentioned learning some basic C++ below. This is in contrast to what I read on places like www.quantnet.com, where I read that people need to be skilled in things like templates, the boost library, etc. Could you comment on this? Thanks.

There are lots of different jobs in finance, and a lot of the trick in getting a job is to match whatever skills that you have. If you are already a skilled C++ programmer, this will help you a lot. However, if you are not, it will take you about three to five years to become one, so if you are looking for jobs in the next six months, it's probably better if don't try to go out for jobs that require you to be a C++ expert, and go out for jobs that just require basic knowledge of the language.

It will take you about three to five years to become a skilled C++ programmer, but getting to the point in which you aren't totally lost with the language can be done in a month or so, and if you have other skills (and if you have a Ph.D. you do have other skills) then that gives you the most bang for the time spent.

 

[twofish-quant]

One other thing. Templates and boost should make programming easier. There are jobs for C++ gurus in finance, but most of what the C++ gurus do is to end up writing libraries and maintaining systems so that people that aren't C++ gurus can do something useful. If you aren't a C++ guru, then you probably aren't going to be one in six months, however, it helps a lot if you spend a month becoming a C++ non-guru

To get a job in finance, you have to be really, really, really good at at least one thing, and decent at other things. The thing that you are really, really, really good at is probably whatever you did for your Ph.D. dissertation. If you have a good headhunter (and you have to look through a few before you find a good one), then part of the conversation will be on how to "sell" the skills that you have. Since you've done instrumentation, have you done microcontrollers and assembly language. If so, then there might be some jobs in algo trading. Done signal processing? GUI? Anything with numpy/scipy?

 

[grapesmoker]

One other thing. Templates and boost should make programming easier. There are jobs for C++ gurus in finance, but most of what the C++ gurus do is to end up writing libraries and maintaining systems so that people that aren't C++ gurus can do something useful. If you aren't a C++ guru, then you probably aren't going to be one in six months, however, it helps a lot if you spend a month becoming a C++ non-guru

To get a job in finance, you have to be really, really, really good at at least one thing, and decent at other things. The thing that you are really, really, really good at is probably whatever you did for your Ph.D. dissertation. If you have a good headhunter (and you have to look through a few before you find a good one), then part of the conversation will be on how to "sell" the skills that you have. Since you've done instrumentation, have you done microcontrollers and assembly language. If so, then there might be some jobs in algo trading. Done signal processing? GUI? Anything with numpy/scipy?

I actually have a fair bit of programming experience with C++ (I've been using it for years), including templates and Boost libraries. My thesis is going to require some signal processing work, but much more image analysis, and I use scipy all the time, so I feel pretty comfortable with that.