Non-academic career options for the theroetical physicist

[daveyrocket]

These are some notes of the observations I've made over the past few months looking for jobs. This might be useful for someone else but keep in mind it's rather focused around my personal situation and preferences. I did my PhD and postdoc in theoretical condensed matter physics, mostly DFT / strong correlations (if you don't know what that is then don't worry about it). I'm looking for a job outside academia, I'm rather bored with physics and I'm desperately trying to stay on the west coast of the United States.

These observations are all biased and anecdotal, so if you read this and thought you found it useful, you're probably wrong.

Materials Science / Engineering - Industry jobs, even entry level ones, seem to overwhelmingly require experimental experience. I managed to interview for the one theoretical solid state physics job I found, and I didn't get it. Or I find jobs where they want someone with more mechanical engineering-type experience, like doing mechanics of materials. It seems there are mechanical engineers to fill those positions, and it's difficult to play a numbers game here because there really aren't many positions like this. (again, had an interview, didn't get the job.)

Oil/gas - The oil/gas jobs that I've seen that I might be remotely qualified for all want experimental experience. Also, you've got to be willing to move to places no one in their right might would want to live in, like Alaska or Texas. On the plus side, you will probably get to travel a lot in certain positions, although I don't really care for that. No idea about salaries. If you're a tree-hugging hippie, then skip this one. In fact, go to the very end of the list.

Engineering

• Civil - Not qualified. There is lots of competition from actual civil engineers. My ex-girlfriend had a degree and experience and good references and she still has a ton of trouble finding a job.
• Mechanical - Theorists might have relevant qualifications (finite element experience) for a small handful of modeling jobs in this field. But not in my case. Experimental experience would be helpful for most positions. It wouldn't be so bad, but there is competition from engineers who already have the right qualifications.
• Electrical - Not qualified. PhD in theoretical physics gives no training in EE.
• Environmental - I think this would be really interesting, but I have no qualifications to get into it. A master's in environmental engineering would go a long way, but no more school for me for a long while.
• Chemical - Even though I minored in chemistry and my background is solid state physics, my experience is not relevant enough because of a lack of laboratory chemistry and experimental methods.
• Geophysics - Don't have related experience. No experimental experience, no training in geology. Sounds really cool though.
• Semiconductor Industry - Nearly all jobs want experimental background. If you're a theorist, you better have worked on silicon.

Information Technology - The main problem I have with doing IT is I had the skills to do it before I even started my bachelor's. The whole reason for going to college in the first place was to get away from these kinds of jobs. Most IT jobs will not have much in terms of upward growth, unless you can/want to get into management. It's probably the easiest field for a physics PhD to be self-employed in though.

• Support - Maybe? I've done tech support before but that was a really garbagety job. Doing high level technical support for people with technical expertise might not be too bad. But this seems like a dead-end career. Might be a good way to pay the bills while studying for the patent bar if nothing else comes along. I have a friend doing technical support for theoretical physicsts, helping them with running codes and interpreting the results of their research. His is one of extremely few jobs like it; there might be as many as a half a dozen jobs like it in the country.
• Management - Lack of management experience is probably a problem.
• Software engineering - More than qualified, but I have back problems that computer use and especially programming makes worse. Plus many of these jobs require long hours. Salaries will be pretty good but probably not increase much with experience compared to other options.
• Web development - Kill me now. I used to consult as a web developer. Web development is the most demoralizing kind of software development, and I will never do it again. Requires long hours typically, and extensive psychotherapy to deal with the fact that you contribute to the moral decay of society every time you write <!--[if IE]> in your code. Contract work is very easy to come by though, so with a little expertise in this area you will not be hurting for work.
• Systems Administration - Viable option. Need to do more research on this one. Seems a bit boring. You probably need to learn a lot about networking. Pay seems average with not a lot of upward mobility. Chances are you will never make $100k in 2011 dollars (although there is a slim chance you might). May require being on call, so if there are problems at 2am you've got to go in and fix them. This might be a great way to get into a different industry, develop contacts and use networking to change out of the IT field into something you'd rather be doing.
• Technical Consulting - Could vary a lot. Sounds stressful and not much fun. Long hours and pay is average. Could be a good option for entrepreneurship. I interviewed with a technical consulting firm but wasn't terribly impressed.

Management Consulting - need an MBA? Don't know much about this. Apparently management consulting firms may give "mini-MBA" training to get you started. Jobs are available, but you can expect long hours and lot of traveling. You do meet a lot of people so you could make a lot of business connections to have a way out when you get sick of it. Unless you have the personality of a bag or rocks, or a theoretical physicist.

Quantitative Finance - Stressful and long hours. Fairly similar in many respects to graduate school. Salaries are very high, better than any other option, but you have to live in NYC. I don't want a job that is similar to graduate school, so no thanks. But with the right personality and circumstance, this could be a very rewarding career.

Defense - No security clearance. Don't want to get it either. Salaries are probably quite high, but I don't know how they compare to patent attorneys or quants. There's a non-compete agreement that if you break you will go to jail for a long time. Also your work will be classified, so if you want to change jobs you won't be able say what you actually did on your resume. Well, you can say what you did, sort of, but you can't go into detail, which only matters if you want to talk to someone who would know what you were talking about. Which they would if you were trying to stay in your field. So this is not a good stepping stone to get to academic research if you care about that sort of thing. If you don't you can always come back to this list after you're done designing bombs that will be used to blow up little brown children. If you haven't developed a conscience by then, the oil/gas field will have an opening for you. Otherwise, see the section on non-profits.

Insurance

• Actuarial - The job market is not large on the west coast. People on forums seem pretty pessimistic about the situation especially for entry level candidates, and especially for physics PhDs, but I do know one person who got in without too much trouble. Salaries start out good, in the 50k-60k range and a lot of room for growth. You have to take tests, which you're probably good at and you might even like if you've gone through a physics PhD program.
• Underwriting - The usual over-/under-qualified condition. Salaries seem low, especially at the entry level, and entry level qualifications are really low, with many not even requiring a 4 year degree. Not many entry level positions available. Could be a good way to get into the industry and make contacts to move up though.
• Sales - I definitely don't have the personality for it and I don't know many physicists who do. Salaries are typically moderately low, in the 30k-40k range. The right positions can pay pretty well though, and if you have the personality for sales you can make hella cash from commissions. But if you had the personality for sales, why the hell did you study physics?

Technical Writing - Might be boring work, but if you like writing and get the right job this could be very interesting. A lot of low salary jobs exist, but I think these have very low education requirements. The right job for a physics PhD could pay quite a bit. Upper end salaries with large tech companies can be fairly high, in the 100k range. Those probably require 10+ years of experience but at least there is upward mobility.

Science Journalism - I have not had an easy time coming up with information on this job. Salaries that I've found range from 35k-70k.

Teaching

• Community college - seems like a garbagety job with poor salary. You get to stay in academics, but you won't have time to do research and so you will probably never transition to a research position. This is career suicide in many ways, IMO. There could be upward mobility into administration though. But because of the way the state budgets have been run into the ground, half your compensation will be awarded in the form of groin kicks (and you're getting them, not giving them). With the current academic job situation, your peers are already all lined up around to block to purchase athletic supporters. But you get to work with students! Many of whom are in a special remedial situation! At least you can get tenure. In most schools anyway.
• 4 year college - lecturer - Few jobs, lots of competition. Salary is decent, around 40k. Up to 100k can be gotten for senior lecturers at large universities. No research support. For a theorist interested in research, this is a viable option since you just need access to computers, and you can build collaborations. But you probably won't have much time to actually do research. With the right motivation for research this could be a stepping stone to a professor position, otherwise it's a dead-end job.
• 4 year college - professor position - Tremendous competition for jobs. Very few jobs, and locations are spread out. Publish or perish. Don't have to publish a lot though. Salary will be decent. You won't have much spare time.
• University professor - If you're going for this job, you already know everything I can possibly tell you about it and you don't need to read this list.
• Corporate Training - Don't know much about this. It could be a decent consulting gig if you have the right connections.
• Tutoring - Low salaries, and you're overqualified. Not only is the hourly wage low you probably will only get a maximum of 20 hours a week. You could freelance easily though.
• English as a foreign language - Apparently there is demand for EFL teachers with technical backgrounds overseas. Could be cool although most jobs are going to be in China and east Asia, so you won't be going to cool places like Buenos Aires. Oh there are jobs in Europe too but probably more competition for those.

Law

• General - No way in hell I'm going to pay for law school tuition.
• Patent Attorney - Passing the patent bar is an option, but not one I will pursue before I need to find my next job. This does seem like it could be really boring too. The salary potential is tremendous though, 2-3 times what you will find in just about any other option. You will probably have to work in a large city, but you should be able to work just about any large city in the country.
• Patent Agent / Examiner - There are ways into the IP field without having the law degree. A job as a patent agent or examiner should be quite good salary-wise, and you might someday get funded to pursue the law degree. Having a degree in electrical, computer, or software engineering would probably be more helpful here than physics, but at least you can still laugh at the civil engineers.
• Forensics - There are jobs that physics PhD's might be qualified for. No idea what the competition is like. These are government jobs: state or federal mostly. Starting pay is decent, 50-60k for state jobs. Federal is probably higher. Maybe there are forensics consulting firms too.
• Enforcement - If you are athletic, you could be an FBI agent and hunt down the aliens that abducted your sister. Pay is good and a background in physics might be rather rare and valuable. Athletic requirements will disqualify most physics PhD's, especially the theorists. You could train, but I think you have to keep up with it as long as you are employed. So if you don't like physical training enough to already be doing it, this is probably the wrong field for you.

Antique Sales - I only put this here because I met a guy at a milonga a couple of weeks ago doing this. He got a PhD in biochemistry, decided it was the wrong field, and now he travels to third world countries, buys up a bunch of antiques real cheap and brings them to the US and sells them. He makes a decent living, enough to travel around the world and learn from tango masters in all the places you would want to go if you liked something interesting like dancing. But I think he was an experimentalist, so he has an actual personality. That's why this job works for him.

Non-profits - Huge variety here and I haven't looked in depth. If you're a tree-hugging hippie who thinks money is evil, this is the industry for you. But it seems a physics PhD will have no special advantage, unlike certain engineering degrees. Salaries will likely be significantly lower than working for a for-profit business or for the government. You get paid with karma instead of money but you like it, you dirty hippie. If you're coming here from the defense industry, you probably need the karma more than the money anyway.

One final comment for job searchers. Network. Network the crap out of the people that you know. I know you hate it but do it anyway. Even if you're a social retard like me. I've only managed to exploit a scant three connections, but it got me four job interviews. (That's not a typo, I've gotten more than one job interview out of one person.) I've also submitted about 100 resumes online, and gotten maybe three interviews. I don't care how bad your social anxiety is, you will still have the lowest effort to payoff ratio through networking. Take an extra dose of your zoloft,* and send that email to that former group member who quit academics in a manner that seemed odd at the time but now you're strangely jealous of him/her.

* Note: I'm not that kind of doctor, so don't take my advice to OD on your SSRI. But the rest of my advice is unquestionably good, obviously.

 

[ParticleGrl]

As someone in a similar boat- phd in high energy theory looking for work, wanting to say in the San Diego area, I can say that my experience has been similar. I would add that management consulting does seem quite willing to hire, but the hours are incredibly long and there is a lot of travel (expect to fly out to a client Sunday afternoon, and fly back in late Friday night, every week). On the upside, its supposed to be a great way to build business connections.

I would also add that community college (at least on the west coast) is a terrible option for the time being. Due to budget constraints, a lot of them are trying to make up shortfalls by hiring adjunct instructors on a per class basis. Teaching a full load, you'd be lucky to scrape by with 20k and no benefits.

I am curious- how did you find technical writing/science journalism type positions? I haven't seen much advertised and would be interested.

 

[Mororvia]

Haha, I know your post was meant to be pretty serious, but it also contained a lot of comic relief. Kudos!

A couple of things for clarification for others who may read:

- If you're interested in patent law, it would could be beneficial to look to start as a patent examiner at the US Patent and Trademark Office or one of the contractor firms that also to work for the USPTO. If the government ever gets its budget shenanigans back in order, USPTO will also likely start paying for folks to go to law school again. Not a bad gig to get paid fairly nice money to be an examiner and then have work pay for your law school as well.

- Just because you have a job that requires a security clearance, that doesn't mean you can list what you do on your resume. You just can't share specifics.

 

[daveyrocket]

Thanks, I incorporated your comments.

ParticleGrl - I found very little on science journalism jobs. I did things like searching for "science journalist" on glassdoor.com. For technical writer, just search for that phrase on a job site, like so:
http://www.indeed.com/jobs?q=technical+writer&l=san+diego%2C+ca

 

[twofish-quant]

I have a lot of experience in oil/gas. Worked five years at a major oil company.

Oil/gas - The oil/gas jobs that I've seen that I might be remotely qualified for all want experimental experience.

There are a lot of coding jobs in oil/gas.

Also, you've got to be willing to move to places no one in their right might would want to live in, like Alaska or Texas.

There are lots of good things about Texas. Austin is a really cool place. Houston... Well. Someone summarized Houston by saying it has all of the problems of a big city and none of the benefits.

If you're a tree-hugging hippie, then skip this one. In fact, go to the very end of the list.

There are a surprisingly large number of tree-hugging hippies in oil/gas. It's not a bad job if you care about the environment and you want to be in a position that you can actually do things to make things less bad rather than just complain about the situation.

One thing about oil companies, is that everyone realizes that the oil is going to run out in the next century and the cheap oil is already gone. The major oil companies are all becoming energy companies, so that whatever technology generates energy in 2050, they'll own it.

 

[daveyrocket]

How are coding jobs in oil/gas different from coding jobs in other industries? Are they similar to the coding jobs in investment banking? What are the hours like and how are the salaries? Is the coding done mostly on modeling problems, solving differential equations, doing finite element calculations, things of that nature? Or are they doing stuff like writing code to automate machinery? What sort of problems are they trying to solve?

 

[Shaun_W]

I consider the environment very important and I'm in oil & gas. Don't assume that just because someone works in oil & gas that they don't give a damn about the environment. When working in oil & gas, you have a lot more say in how the environment is treated than if you chain yourself to trees whenever someone is going to build a new runway and write lots of blog posts.

 

[lisab]

I consider the environment very important and I'm in oil & gas. Don't assume that just because someone works in oil & gas that they don't give a damn about the environment. When working in oil & gas, you have a lot more say in how the environment is treated than if you chain yourself to trees whenever someone is going to build a new runway and write lots of blog posts.

Same can be said for many industries.

Years ago when I worked for a paper company, there was a woman who got a lot of media attention for sitting in a tree and not coming down, to save the tree.

At the same time there was a guy I worked with, a chemist, who was obsessed with finding a low-or-no-effluent way to make paper. He worked on this even on his own time!

Twenty-plus years later, several of that chemist's ideas have been successfully implemented. The tree that woman sat in, it was cut down (albeit by vandals).

Real change happens from the inside, often.

 

[Andy Resnick]

These are some notes of the observations I've made over the past few months looking for jobs.

This is a remarkably complete post, and based on my experience (I got my PhD in 1997 and have been in several of the sections) very accurate. Thanks!

Of course, nobody has commented on the obvious- earn a degree doing something that can be easily transferred to an industrial setting. Theoretical physics ain't it.

 

[ParticleGrl]

Of course, nobody has commented on the obvious- earn a degree doing something that can be easily transferred to an industrial setting. Theoretical physics ain't it.

A point that should be regularly repeated. As much as it seems like common sense, I honestly wish any of the people I relied on for advice in my academic years had made this point. Yes, you should follow your dream, but temper it with reality.

 

[twofish-quant]

Of course, nobody has commented on the obvious- earn a degree doing something that can be easily transferred to an industrial setting. Theoretical physics ain't it.

I think it's not that simple...

On the one hand my astrophysics degree has been really useful for my finding an industrial job. But on the other hand, it's because I've spent a lot of time in areas other than on assigned homework, and it happens that the type of theory that I did involved spending eight hours a day for five years on front of a computer coding...

 

[twofish-quant]

As much as it seems like common sense, I honestly wish any of the people I relied on for advice in my academic years had made this point. Yes, you should follow your dream, but temper it with reality.

It's more complicated. I've had much more success in following my dreams, and ignore people that wanted me to be "realistic." One thing that happened in my undergraduate years was that I was extremely curious about things other than physics, so rather than study only physics, I spent a fair amount of time studying things like economics or C++ or French postmodern philosophy or Chinese constitutional law. This really hurt my grades, it hurt me when I went into graduate school and pretty much eliminated any chance that I had of being faculty.

But I ended up doing quite well.

I suppose whether you should follow your dreams depends on what your dreams are, and where they come from. One of the things that I was able to do was to do some historical detective work. Why do I believe this? Where did my dreams come from? A lot of it came from my parents and teachers, but that's just the start of the mystery. Where did *they* get their dreams from?

One day I was in the library reading a book on 18th century Chinese philosophy, and I came upon a philosopher (Dai Zhen) and reading about him was bizarre because by some weird coincidence, he happen to believe almost exactly what I believed. It took me a few weeks to realize that it wasn't a coincidence. He happen to come from the same part of China as my parents, so *he* was in large part responsible for brainwashing me.

Something that helps a lot is to figure do the "been there, done that thing." The dream of every upwardly mobile Chinese family in the 19th century was to pass the Imperial examinations. There was one major problem too many degrees, too few jobs, and it's interesting to see what people did.

Figuring out where your dreams came from helps you to figure out how to revise them.

 

[twofish-quant]

How are coding jobs in oil/gas different from coding jobs in other industries? Are they similar to the coding jobs in investment banking? What are the hours like and how are the salaries? Is the coding done mostly on modeling problems, solving differential equations, doing finite element calculations, things of that nature? Or are they doing stuff like writing code to automate machinery? What sort of problems are they trying to solve?

Hard question to answer. If you ask what the life of a HEP physics professor is like, then it is easy, because there are about a dozen or so hired each year, and you can take a look. Asking what the life of a coder is like is much more difficult. The good news is that there are a lot of jobs. The bad news is that because there are a lot of jobs, it's hard to generalize.

The analogy I like to use is that coding is a specialized form of "writing." Knowing that someone in a company "writes stuff" tells you very little at all about what they do.

 

[Sankaku]

But the rest of my advice is unquestionably good, obviously.

I am not looking for employment and don't have much useful to add; however, I enjoyed your cynical sense of humour and wanted to say that technical writing jobs can be very diverse if you have any skill at the written word (which it looks like you do).

It may not be demanding on your theoretical skills but getting your foot in the door can lead to other positions as well.

 

[Ryker]

Hmm, reading this makes me kind of wonder, though. If people going into theoretical physics are close to mathematicians in terms of courses they have to take, and the amount of maths they know, why don't PhD's in theoretical physics have the same avenues open as the mathematicians do. I'm reading more and more how desired the latter are in industry, so what are theoretical physicists lacking in comparison, and how hard would it be to remedy those deficiencies? My guess is that not much, so unless my guess is wrong, the outlook for theoretical physicists shouldn't be that bleak.

What am I missing?

 

[daveyrocket]

What sort of avenues do mathematicians have open to them that you have in mind? I've uncovered a handful of jobs looking for statisticians. For someone like me who is avoiding software jobs, it's pretty bleak.

I consider the environment very important and I'm in oil & gas. Don't assume that just because someone works in oil & gas that they don't give a damn about the environment. When working in oil & gas, you have a lot more say in how the environment is treated than if you chain yourself to trees whenever someone is going to build a new runway and write lots of blog posts.

That's a good point, and I'd add some text to that effect if I could edit the topic.

 

[maverick_starstrider]

Well as a person in Computational Condensed Matter myself this is certainly off-putting. I have a Masters and am in my first year of PhD (done in about 3). I personally like to temper my physics explorations with Computational know-how (Parallel-programming, exact diagonalization, finite element, etc.) and I also had a minor in Applied Math in my undergrad. Once I get out I don't know if I have any strong interest in continuing in physics (although if the option was open I'd certainly pursue it but I'm just not married to the idea). I'd be perfectly happy working in some sort of computational or mathematical modeling situation, scientific or engineering computation, etc (i.e. same toolbox different subject). I'm wondering to what extent you feel that your experience overlaps with my own. I also have no experimental.

 

[daveyrocket]

Yeah my experience is very similar to yours. My two main problems are my location restriction and my aversion to coding jobs. There seem to be rather few mathematical modeling jobs in my area. There are some but not many.

None of the condensed matter specific stuff you will do will be relevant to any job in industry save a very select few. If you don't think you're going to have a strong interest, why are you doing the PhD?

Edit: oh, the finite element stuff could be useful.

 

[maverick_starstrider]

Yeah my experience is very similar to yours. My two main problems are my location restriction and my aversion to coding jobs. There seem to be rather few mathematical modeling jobs in my area. There are some but not many.

None of the condensed matter specific stuff you will do will be relevant to any job in industry save a very select few. If you don't think you're going to have a strong interest, why are you doing the PhD?

Edit: oh, the finite element stuff could be useful.

The same reason anyone does physics I suppose, the intellectual challenge, the weighty ideas, avoiding the job hunt, etc. I'm computational so it all depends on what is meant by "coding" work, numerical computation is my bread and butter and area of speciality (i.e. algorithm design that exploits computational knowledge to do math faster/better). However, I really have no interest in software development of like apps or some such (nor do I really have the expertise). But ultimately I'd be perfectly fine with bringing my physicist toolbox to new fields, I worked in my undergrad writing scripts to do image segmentation for example (basically writing codes to pull information out of pictures, specifically MRI scans). So has your experience been that those jobs don't really exist or that you've seen them but didn't feel qualified to apply? It's been my experience that all I really want is a good puzzle put in front of me and room to work, whether it's critical phenomena or targeting demographics for marketing toasters or some such I don't know if it really matters so much.

 

[creepypasta13]

Of course, nobody has commented on the obvious- earn a degree doing something that can be easily transferred to an industrial setting. Theoretical physics ain't it.

A point that should be regularly repeated. As much as it seems like common sense, I honestly wish any of the people I relied on for advice in my academic years had made this point. Yes, you should follow your dream, but temper it with reality.

From what I've gathered, that's not true for theorists with at least some computational skills

 

[twofish-quant]

My guess is that not much, so unless my guess is wrong, the outlook for theoretical physicists shouldn't be that bleak.

I think it depends on the particular theoretical physicist. I've had a wonderful time with my degree.

One reason I think I've done well is because of my parents and my teachers. My father had the bad luck of joining the losing army in a civil war just before they lost, so he spent his life moving from career to career and country to country, so this seemed "normal to me." One of the most influential people in my life was my Latin teacher who also moved from career to career. He also was in the military, became an international lawyer, went into the world of politics, then eventually ended up teaching high school.

 

[ParticleGrl]

From what I've gathered, that's not true for theorists with at least some computational skills

Look at the original-post- a physicist with some computational skills having trouble finding jobs that match his qualifications. Lots of fields graduate people with computational skills (all forms of engineering, many mathematicians, computer science,etc), and theoretical physics doesn't seem to give you anything to give yourself an edge over those other guys. At literally EVERY interview (even for bar-tending and waiting tables!) I've gotten, the question "why should we hire you when these other guys have more directly relevant experience?" has come up. Further, jobs in science that want only computational experience and no bench work seem to be fairly rare.

I suppose if you are cool with IT work or a programming/development job, you can easily get it. Of course, you need neither a bachelors degree nor a phd to pursue those jobs- which can be frustrating.

Apart from that, a lot of the computational jobs are highly region dependent. If you aren't going to relocate to a finance center, for instance, you are going to have trouble finding a finance job.

 

[ParticleGrl]

I'm reading more and more how desired the latter are in industry, so what are theoretical physicists lacking in comparison, and how hard would it be to remedy those deficiencies? My guess is that not much, so unless my guess is wrong, the outlook for theoretical physicists shouldn't be that bleak.

I'm not sure that mathematicians are that desired in industry? Maybe the mathematicians who are closer to the CS field? I only have anecdotal experience, but about 10 fairly good friends from undergrad now have phds in math, and about 4 more from graduate school.

Roughly half are in postdocs. With one exception the rest all had extremely difficult times finding industrial work (a year or more of searching, one guy is now getting a masters in actuarial science, another is going to nursing school). The exception was a guy who did applied math and met his future employer at an industry workshop.

 

[ParticleGrl]

I think it depends on the particular theoretical physicist. I've had a wonderful time with my degree.

And you also had the good fortune to graduate at a time when dot-coms were snapping up technical talent as fast as they could. Once you get that one job, its much easier to get others, and when the labor market isn't as glutted, people are willing to hire employees who aren't quite perfect matches.

I think when there is a shortage of workers, you hire the trainable people. When there is a glut of workers, trainable just means not-qualified-yet.

 

[daveyrocket]

The same reason anyone does physics I suppose, the intellectual challenge, the weighty ideas, avoiding the job hunt, etc. I'm computational so it all depends on what is meant by "coding" work, numerical computation is my bread and butter and area of speciality (i.e. algorithm design that exploits computational knowledge to do math faster/better). However, I really have no interest in software development of like apps or some such (nor do I really have the expertise). But ultimately I'd be perfectly fine with bringing my physicist toolbox to new fields, I worked in my undergrad writing scripts to do image segmentation for example (basically writing codes to pull information out of pictures, specifically MRI scans). So has your experience been that those jobs don't really exist or that you've seen them but didn't feel qualified to apply? It's been my experience that all I really want is a good puzzle put in front of me and room to work, whether it's critical phenomena or targeting demographics for marketing toasters or some such I don't know if it really matters so much.

The jobs exist, just not in numbers. Unless you don't care where you live. Even then there aren't that many, but enough to play the numbers game. I've applied for nearly every one I found that meets my area restrictions, and the only interviews I got were from the networking that I did.

Having experience in something medicine-related could help a lot. I see a lot of listings for someone knowledgeable in statistics and medicine. So your background with MRI's could set you apart in a way that you might not have guessed. But make sure you actually know something about statistics, not just what they teach you in the math methods for physicists class.

Personally, if I could do it over again, I'd quit physics after getting my master's and switch to getting a master's degree in either statistics, mechanical engineering, or environmental engineering. Those fields all seem more relevant, and more interesting.

From what I've gathered, that's not true for theorists with at least some computational skills

I should have written a bit more about software jobs. The problem is, for most positions you're less competitive than someone who's had a couple of years experience in industry because while you were learning QFT and determinant Monte Carlo, they were learning about things like the software development life cycle, AJAX and whatever other hot new technologies are relevant to that industry. The current economy is where it really hurts to have a much more general training rather than the specific training on what's needed, or even part of the specific training that's needed.

Further, jobs in science that want only computational experience and no bench work seem to be fairly rare.

Yeah, and from what I've found, those are even worse because none of them are physics. They want chemistry or pharmaceutical or biology or engineering backgrounds. Okay, if you're in condensed matter you might have some background relevant to quantum chemistry (but not if you're solving model Hamiltonians). But you're not going to be better off than a quantum chemist applying to that position, and there aren't going to be a lot of positions out there so you can just fire off 50 resumes and expect to get a call or two.

 

[creepypasta13]

Personally, if I could do it over again, I'd quit physics after getting my master's and switch to getting a master's degree in either statistics, mechanical engineering, or environmental engineering. Those fields all seem more relevant, and more interesting.

Hmm, that's an interesting idea. I was always indecisive between aerospace engineering and physics, but chose to accept an offer for a Phd program in Physics for the love of the subject despite the much better opportunities in AE. I never considered getting an MS in Physics and then get an MS/Phd in AE though I should re-consider

Look at the original-post- a physicist with some computational skills having trouble finding jobs that match his qualifications. Lots of fields graduate people with computational skills (all forms of engineering, many mathematicians, computer science,etc), and theoretical physics doesn't seem to give you anything to give yourself an edge over those other guys. At literally EVERY interview (even for bar-tending and waiting tables!) I've gotten, the question "why should we hire you when these other guys have more directly relevant experience?" has come up. Further, jobs in science that want only computational experience and no bench work seem to be fairly rare.

I suppose if you are cool with IT work or a programming/development job, you can easily get it. Of course, you need neither a bachelors degree nor a phd to pursue those jobs- which can be frustrating.

Apart from that, a lot of the computational jobs are highly region dependent. If you aren't going to relocate to a finance center, for instance, you are going to have trouble finding a finance job.

The OP said he doesn't want to work in Defense. After getting my Phd, I planned on a career in Defense or DOE that requires a clearance. From what others have told me, lots of Physics PhD's get jobs in defense, though I never asked if they were theorists or experimentalists. I have a BS in Physics (will start the phD program soon) and got a few job interviews, but they were mostly either Defense-related or programming-related. But I never got the programming jobs because my skills aren't nearly as good as CS majors.

 

[twofish-quant]

Look at the original-post- a physicist with some computational skills having trouble finding jobs that match his qualifications.

If you are looking for a computer job, but you don't like or tolerate coding, you are going to be in for some difficulty.

Lots of fields graduate people with computational skills (all forms of engineering, many mathematicians, computer science,etc), and theoretical physics doesn't seem to give you anything to give yourself an edge over those other guys.

Depends on the type of theoretical physics. My first computer job was with an oil company that had a ton of FORTRAN code that needed to be baby-sat, and I just happen to be one of the few people that had FORTRAN on my resume. Also, I've done grid computing and I've worked on algorithmic systems that are tens of thousands of lines long.

The other thing is that even with a soft job market, you need only be "good" and not "perfect" to get a job.

At literally EVERY interview (even for bar-tending and waiting tables!) I've gotten, the question "why should we hire you when these other guys have more directly relevant experience?" has come up.

And the answer that I usually give is that I have more directly relevant experience than the other guys, even if the interviewer doesn't realize it. This involves a bit of sales and marketing. It also involves convincing myself that I really should get the job. If I can't convince myself that I'm not the best person for the job, then I'm not going to convince anyone else of it.

Apart from that, a lot of the computational jobs are highly region dependent. If you aren't going to relocate to a finance center, for instance, you are going to have trouble finding a finance job.

On the other hand it's less bad than academia in that if you want to work in field X, you can usually figure out where the hot spots are. I suppose that's one of the advantages of having immigrant parents and an immigrant spouse. If your parents moved half way around the world to get a job, moving to another city isn't that big a deal, and if your spouse moved half way around the world to go to school, then relocation isn't a bad deal either.

 

[twofish-quant]

The jobs exist, just not in numbers. Unless you don't care where you live. Even then there aren't that many, but enough to play the numbers game.

The other thing is that it's really hard to tell how much is "Ph.D." and how much is "generally bad economy." It's a struggle to get a job in this economy, but I've got the sense that as tough as it is, Ph.D.'s are still doing better than people in other fields.

Personally, if I could do it over again, I'd quit physics after getting my master's and switch to getting a master's degree in either statistics, mechanical engineering, or environmental engineering. Those fields all seem more relevant, and more interesting.

But I'm not convinced that they are more likely to get you a job.

The problem is, for most positions you're less competitive than someone who's had a couple of years experience in industry because while you were learning QFT and determinant Monte Carlo, they were learning about things like the software development life cycle, AJAX and whatever other hot new technologies are relevant to that industry.

This is one of the "stupid things" that I did as an undergraduate that helped me a lot afterwards. I did my undergraduate project on educational software development, and while I was working on my Ph.D., I did lots of stuff that build on top of the CS experience. So when I got out, I could put that I had a decade of software development experience, and this was also in the middle of the dot-com bubble.

The current economy is where it really hurts to have a much more general training rather than the specific training on what's needed, or even part of the specific training that's needed.

Not sure that's true. One other thing that helped me look for work was reading sales and marketing literature. Often the client/customer/employer really doesn't know what they want, so what I do going into a sales situation is to not take the employers requirements at face value.

This is where "follow my dream" worked. I was under *enormous* pressure to focus and to do nothing outside of physics, but I was able to resist that. One reason I did resist this was that the people that were important in my life encouraged me not to be a specialized tool.

The problem with specializing is that you are doubly dead if you specialize on the wrong thing, and it's going to end badly if you specialize in exactly the same thing that everyone else specializes in.

 

[Andy Resnick]

The same reason anyone does physics I suppose, [...], avoiding the job hunt, etc.

This says more than I ever could.

 

[elkement]

Hi daveyrocket,

great post! Your list of options in IT is accurate and complete. Although I work in the IT industry I fully agree to your slightly negative overall conclusion.

I think an IT job is only rewarding if you manage to find an area of expertise that allows you to pick engagements / projects that match your preferred way of working.

In any corpration IT is considered basic infrastructure that should just work - you only get visibility if something breaks. IT tries to market itself in terms of 'enabling business strategy' etc., but end users and management rather compare IT to plumbing. And nearly any IT department I know is understaffed. People operate in permanent emergency mode and need to meet insane deadlines (deadlines often related to politics not to real requirements) - this is true both for the support guys and for IT managers.

What ParticleGrl says re management consulting is also true for technical consulting: As an employed consultant you typically need to meet goals in terms of number of billable hours. A consulting manager is responsible to shuffle human ressources and make sure that your utilization is high enough - e.g. by selling you to a client for 5 days a week ... which results in this traveling lifestyle. This 'utilization pressure' was one of the main reasons I did not work as an employed consultant any more.

I agree that IT allows you to launch a career as a self-employed developer or consultant rather easily. But I think you are facing the same challenges: You would probably want to avoid both 24/7 support for small companies or working in long-term projects living the traveling consultant lifestyle. Besides the market for such large projects is (at least in Europe) dominated by recruiting compnies / agencies that serve as intermediaries between freelancers and clients, which actually reduces the theoretical 'freedom'.

There are still attractive niches in terms of subject and work style (IT forenscis for example, working as a reviewer and investigator in general, but not in IT operations and long-term projects.). But I believe these opportunities become available after you have gathered some experience - on a very specific subject or in an industry sector. This typically takes a while and - you name it - networking, talking to the right people, market yourself etc.

Good luck,
Elke

 

[Ryker]

What sort of avenues do mathematicians have open to them that you have in mind? I've uncovered a handful of jobs looking for statisticians. For someone like me who is avoiding software jobs, it's pretty bleak.

I'm not sure that mathematicians are that desired in industry? Maybe the mathematicians who are closer to the CS field? I only have anecdotal experience, but about 10 fairly good friends from undergrad now have phds in math, and about 4 more from graduate school.

Roughly half are in postdocs. With one exception the rest all had extremely difficult times finding industrial work (a year or more of searching, one guy is now getting a masters in actuarial science, another is going to nursing school). The exception was a guy who did applied math and met his future employer at an industry workshop.

I was just thinking of jobs in insurance companies, companies that develop new models in whatever field they're dabbling in etc. My impression might have been wrong, but even the BLS website says the number of jobs for mathematicians is going to grow much faster than the average. Now I know I can't go from that alone, but I guess my reasoning was that they couldn't be that far off.

 

[daveyrocket]

Yeah, those jobs exist but again not in huge numbers. A lot of the ones I see that I think look very interesting want some significant background in statistics, with experience using statistical software such as SAS. You may or may not have that kind of background from your physics training. I actually never took a course in probability or statistics. Some things like probability distributions and stuff used in QM I know very well, but other things like conditional probability I never learned. And I know some stuff about statistics from statistical mechanics, but I've never done things like t-tests or chi squared tests, I don't know how to do an ANOVA and I've never used R or SAS.

If you spend time during your PhD to develop skills that will actually be useful outside of academics then you could make yourself very competitive for jobs. And the information is easy to get -- search around for job listings and find ones that you might like to do, then look at the requirements and figure out how you can develop those skills. I'm not saying you shouldn't do a PhD in physics, but of course you might realize the best way to get one of those jobs would be to switch from a PhD in physics to e.g. a Master's or PhD in statistics.

If you spend your time during your PhD with your head in the sand about job related stuff and don't worry about it until you graduate, then you might find that you don't have time to pick up necessary or useful skills for non-academic jobs that look very interesting. And the worst thing is it will be obvious that you could have easily picked up those skills along the way if you had known. Don't count on your physics adviser to teach you those skills, or even know what skills will be useful.

 

[twofish-quant]

In any corpration IT is considered basic infrastructure that should just work - you only get visibility if something breaks.

Then you get into the wacky world of marketing.

Imagine an ad on TV, in which you are invited to go eat dead cow parts and then crap in the bathroom after you are done. Doesn't sound pleasant.

However, if you go to a five-star restaurant, that's what you will be doing. The dead cow parts may be Waygu steak, and the bathrooms are going to be really nice with fresh flowers.

In the corporate world, the term "IT" means "eating dead cow parts." People that "do IT" end up running cables and installing Microsoft Office.

IT tries to market itself in terms of 'enabling business strategy' etc., but end users and management rather compare IT to plumbing.

True, which means that if you want a functioning IT department, the first thing that you want to do is to not call it IT. When you say IT, you think "dead cow parts." One fun marketing term is anything with the term "solutions." There are a lot of other bureaucratic tricks that you can use in the wild and wacky world of corporate politics. You can divide IT responsibilities among several groups so that none of them are IT.

As an employed consultant you typically need to meet goals in terms of number of billable hours.

Also a lot of "technical problems" turn out to be political problems. You have an IT department that is understaffed and underpaid, and so they get screamed at for being idiots. At this point, desperate manager brings in IBM who then brings outside consultants that get paid three times as much money for the same work and who aren't subject to the silly bureaucratic rules that keeps local IT from doing the same thing. You can make money in this situation, but you do realize that local IT will hate you, and your only solace is that you'll be gone before they hate you enough to do physical damage to you.

A consulting manager is responsible to shuffle human resources and make sure that your utilization is high enough - e.g. by selling you to a client for 5 days a week ...

Also depending on the company, you may turn out to be a burger flipper, or not. Technical consultants get treated very well at IBM. However, at Accenture and any place that grew out of accounting, you'll be given orders by managers that are pretty much totally clueless.

One good news is that if you have contacts, you can actually to into business for yourself. However, after 2007, this isn't as popular as it was before.

Besides the market for such large projects is (at least in Europe) dominated by recruiting compnies / agencies that serve as intermediaries between freelancers and clients, which actually reduces the theoretical 'freedom'.

In the US, there is one big thing that keeps people from doing independent consulting and that is health insurance. If you or your spouse have any pre-existing condition, then you can't get health insurance at anything decent, which means that you can't run your own business.

On the other side, recruiting companies / agencies in the US are often mom-and-pop shops, so there are a bunch to choose from.

This typically takes a while and - you name it - networking, talking to the right people, market yourself etc.

And if you do it right, it helps to have the letters Ph.D. on your business card.

 

[twofish-quant]

Yeah, those jobs exist but again not in huge numbers. A lot of the ones I see that I think look very interesting want some significant background in statistics, with experience using statistical software such as SAS.

Then get it. Go to amazon, by a book on SAS, and teach it to yourself. If you can learn quantum field theory, then with about a month of independent reading, you can do SAS better than 90% of the people that are applying for those positions.

There are positions that require total expert knowledge of SAS, but there you'll find that most of the statistics positions are "burger flipping" that just needs someone to run some data through a program. You can pick that up pretty quickly.

You may or may not have that kind of background from your physics training.

If you don't know something then learn it. It so happens that I've used both SAS and R. It's not QFT. Now convincing someone that you have enough SAS and R ability to get the job done takes a bit of doing, but all you really have to do is to take SAS run some experimental data against it, and boom, you have something on your resume.

I actually never took a course in probability or statistics

Neither have I. I've also never taken a course on C++, and I've only taken two courses in computer programming in my entire life. I hate taking courses. I like learning new stuff.

One reason I hate some courses is that they are often too slow. If you have done quantum field theory and you take your typical course in probability and statistics for social science majors, you will get extremely bored and annoyed after the first week, because the courses is moving just way too slow.

And I know some stuff about statistics from statistical mechanics, but I've never done things like t-tests or chi squared tests, I don't know how to do an ANOVA and I've never used R or SAS.

R is freeware, and you can google for t-tests and chi-squared.

If you spend time during your PhD to develop skills that will actually be useful outside of academics then you could make yourself very competitive for jobs.

ANOVA is just basic algebra. If you can figure out how to do QFT, you can do an ANOVA. It took me about two days to learn how to do it. Chi-square and T-tests are also things that will take you at most a week to learn.

I'm not saying you shouldn't do a PhD in physics, but of course you might realize the best way to get one of those jobs would be to switch from a PhD in physics to e.g. a Master's or PhD in statistics.

No...

It's like telling someone that they should go to cooking school so that they can work at McDonalds. Most of the jobs that require R and SAS experience are for companies that need a burger flipper so that they can run the data. If you are a physics Ph.d., you can likely learn that stuff on your own.

If you spend your time during your PhD with your head in the sand about job related stuff and don't worry about it until you graduate, then you might find that you don't have time to pick up necessary or useful skills for non-academic jobs that look very interesting.

And it's not too late after you get your Ph.D. If there are a ton of jobs that require SAS experience, go out and learn SAS. You won't be the world expert in SAS, but my guess is that the world expert in SAS is busy doing something else, so they may have to hire you because no one else is available.

And the worst thing is it will be obvious that you could have easily picked up those skills along the way if you had known.

And since you already have your Ph.D., it's too late since there is this rule that you can't learn anything new once you have your Ph.D. (that was sacarsm).

1) there is this thing called google and this other thing called amazon. I hear that you can get information on how to run SAS.
2) yes it's a bummer to know that you could have done something different, but that's life. One thing to remember is that there are times in which you have new jobs with new skills in which no one *could have known* what skills would be useful. If five years ago, it was obvious that there would be a ton of jobs in SAS, then everyone would take SAS courses, and we'd have a glut of SAS programms.

Don't count on your physics adviser to teach you those skills, or even know what skills will be useful.

.

But that's cool. Ultimately its your dissertation and your life. You physics adviser may be the world expert in particle physics, but he is an *adviser* and his advice in auto repair or pastry cooking may not be that good. Now it turns out that I know something about SAS since my wife took a class in it, and while she was doing it I looked at her books, and she has friends that do work as SAS burger flippers.

But I'm trying to understand what the basic problem is...

I got my Ph.D. because I wanted to learn new stuff. I like learning new stuff. The fact that most of the stuff that I know is going to be useless or obsolete in five years is cool.

Again maybe it's because of my father. I have a lot of admiration for my father because he was able to do and learn whatever it took to keep us fed. He was trained as a mechanical engineer, but there was a point at which the only job he could find was to teach air condition repair. So he taught himself how to fix air conditioners. He got a job at a community college. Then in 1978, Radio Shack came out with the TRS-80 Model I. There was a big demand for people that could program in BASIC, so he taught himself how to do that, and became the lead instructor in microcomputers.

 

[twofish-quant]

Lots of fields graduate people with computational skills (all forms of engineering, many mathematicians, computer science,etc), and theoretical physics doesn't seem to give you anything to give yourself an edge over those other guys. At literally EVERY interview (even for bar-tending and waiting tables!) I've gotten, the question "why should we hire you when these other guys have more directly relevant experience?" has come up.

There are two standard answers that I pull out.

1) "broad diversity" - you are hiring bottlecap washers and every one in your company has a masters degree in bottlecap washing. You should hire me because I *don't* have a degree in bottlecap washing. I have a Ph.D. in astrophysics which means that I can come up with some new ideas for bottlecap washing like ...

2) "I learn quick" - if you hire someone that knows bottle cap washing then they are going to freeze when the bottle cap industry changes. Now if you hire me, I can learn new techniques for bottle cap washing really fast. For example, six months ago, I knew *nothing* about bottle cap washing, and as you can see from the test that you gave me, I've taught myself the basics at it...

One thing to remember is that if you have gotten to the point where someone is doing an interview, there is *something* that they see in you that has them talk with you, and you can push on whatever that thing is.

For example suppose I'm looking for a job being a waiter... The first thing that I would do is to go down to the library and read some books on restaurant management, and luck would have it, I was bored one day and I actually did read some books on restaurant management. So the pitch would be...

- look I need a job to tie me over until I find something better, but that's fine because no one in the restaurant business gets a job as a waiter for a lifetime career. What you probably care about is that I'll show up to work on time, pull my shift, and take orders from you and provide a great experience for your customers, and I've had plenty of experience working at that being a teaching assistant. Dealing with a difficult customer is like dealing with a difficult student that demands a higher grade, and I've done that.

Also I've read that one of the important skills for a waiter is to be able to balance plates, and I've been practicing that for the last three weeks. You can had me that tray over there and test how well I do that.

Also for bartending, it turns out that one of the big things is compliance with liquor laws, so at the interview, I can mention that along with memorizing the basic cocktails, I've also read about the procedures to follow to inventory the liquor and that I've already taught myself the relevant liquor laws. Also with the down economy, a lot of the customers are likely to have job problems, and I can provide a sympathetic ear for that...

It's all sales and marketing... And if you can figure out quantum field theory, you can figure out waiting tables and job interviews. Also it's useful to put yourself in the right mindset, I get bored easily so one thing that I've done when I'm at a restaurant is to look very closely at the wait staff, and see what's going on. Part of "figuring out the mysteries of the universe."