Near the End of A PhD and Have No Job

jasonRF

I will have to agree with all of the above. I know quite a few folks who went to work at FFRDCs with backgrounds in high energy physics, nuclear physics, etc., and are doing work side by side with engineers. I also have a PhD (but in engineering) and have interviewed a lot of potential employees. I look for a) smart people with strong technical background, who b) are interested in our company and have a clue about us, and who are c) enthusiastic about what they have worked on in the past and can d) clearly communicate their work to an audience of people who do not share their specialty. We have hired quite a few Physics and Math PhDs to work for us, and in general they work out quite well for doing engineering work, and bring new angles to problem solving that we wouldn't get if we only hired engineers.

I must say that particular knowledge can matter, though, especially if our current work is more rapid, short-term types of projects that don't allow for the time it takes for non-specialists to get up to speed. My group does a lot of image and signal processing, almost all of which is based upon statistical theory. Our last Physics PhD I interviewed didn't get hired, primarily because he didn't know probability and statistics sufficiently well to be useful out of the box (I must say that this is by far the biggest problem we have with physicists - physics departments could greatly increase the marketability of their students by forcing them to take a one-year probability and mathematical statistics sequence at the senior level. Add in a stochastic processes class and a Fourier analysis course and they would be great hires!). If you have a clue about probability and statistics you need to get that into your resume somehow. And if you didn't take coursework on it, be prepared to be quizzed. If it were me, i would want you to know:

prob: basic probability, conditional probability, expectation, something about gaussian random vectors and linear transformations of gaussian random vectors.

estimation: least-squares, maximum likliehood, and know what the Cramer-Rao bound is and why it matters. It would be a bonus if you know mean-square estimation as well (aka minimum mean square error estimates).

hypothesis testing: Neyman-Pearson lemma and likelihood ratio tests. Composite hypothesis testing would be included here, when you do not know noise levels apriori you have to estimate them (engineers call this generalized likelihood ratio tests).

This stuff isn't hard and I wouldn't ask you for detailed derivations, just basic ideas that made it clear you understood the idea.

best of luck.

jason

ParticleGrl

Is there any good way to discover if a company is doing more short-term projects? I've grown frustrated with employers telling me "we need someone ready to go out-of-the-box." Is there a good way to find companies with longer-term projects?

jasonRF

I don't know of any good way. The best way I can think is to look at places that have a long history of hiring physics/math phds. When I was in grad school I was in a plasma physics group (was engineering, but not so useful ...) and many of our graduates interviewed with places like APL, Lincoln Lab and other ffrdcs and got offers. Granted it was the late 1990s before the dot com bubble burst. Anyway, often the job was to do "system analysis" which seemed like a topic that no one learns in school and they just wanted smart people that they could train. No, I do not know what system analysis really means, but if they still hire for those kinds of jobs then you already have the qualifications you need.

Some of the military labs may be worth a look, too. AFRL and ARL come to mind.

best of luck,

jason

JDGates

You can find out a good amount with some serious Googling. If you have still have access to an academic library (even by just walking in) then LexisNexis can be useful. I also had some luck using alumni network searches.

To take my company as an example, it's true that you won't be able to find out anything about ~95% of the specific projects we're working on; what gets made public is nonrepresentative. However, you should be able to find out who our major sponsors are and what sets us apart from other companies working for the same sponsors. You should also be able to figure out why we like science and engineering PhDs.

It does take some work and I don't guarantee that it will always work, but putting some time into researching companies does pay off, at least in my experience. Being able to (intelligently) use certain keywords and phrases before my interviewers did appeared to work to my favor, anyway.

It really depends. If it's an engineering-type job, then I'd think that lacking specific technical skills (e.g., experience with some software package or statistical method) would be more detrimental; jasonRF appears more qualified to speak to this than I am. If it's a more analytically-oriented job, then they may be more interested in your problem solving skills, and let you learn specifics on the job (this is the attitude where I work), making it more worthwhile to spend time applying.

Absolutely. My undergrad physics department did a great job of mixing stats in with the upper-level labs, and I took some stats on my own, but I still wish I'd had more. I don't know if it would have helped in my job search, but it certainly would've made doing my job easier at times.

Aaronvan

Try the National Security Agency. They hire a lot of people with your quals. Having a foreign language is a plus, too.

twofish-quant

This is highly field dependent, and not applicable to finance. Googling for what my company is pretty close to useless, and our webpage doesn't give much in the way of information about what Ph.D.'s do. There are people in the company that spend their days googling for information and if they find an information leak, they'll plug it.

We do have lots of "meet and greets" and face to face recruiting sessions

And we are one of the more open companies. We have a web page. There are hedge funds that don't have a web page.

I should point out that this is one part of the culture that I don't like, and I'd like for us to be more open about what we do since a lot of it is mathematically and scientifically cool. The interesting thing is I don't think that anyone is *against* that idea. It's just that to change the standard operating procedure of saying nothing involves a lot of fighting inertia.

That helps a lot in the finance interview. Knowing what a greek, bond duration, put-call parity is helps a lot, and it will take you two days to pick up the jargon. I had one interview in which the recruiter mentioned was at in fixed income, so I went to the book store and started reading. I wasn't a bond expert, but it helps to understand the question that someone was asking me. Also a lot of it is easier.

greek = math derivative
delta = first derivative
duration = scaled first derivative
gamma, convexity = second derivative

Also it depends on how many people apply. If you have one spot and three good candidates, the person with experience in a specific package might get the job, because there is nothing else to decide between the three. But it's very much something of a lottery ticket.

Curiously, I got me first job because I was a FORTRAN 77 guru. There were Java experts all over the place, but very few people that knew FORTRAN 77. Some people collect antique wine. Some people collect antique cars. I've gotten jobs knowing out to deal with antique code.

This gets into the danger of following the crowd. One of the few things that you can do to maximize your winning in the lottery is to not pick the same numbers as everyone else. That's why questions like "what should I study" are somewhat self-defeating, because if everyone studies something, you'll get a bubble, and then people won't make money at it.

hjlim

Hi,
What about taking some, say, computer science courses in your school? If you are willing to stay in school for 1~2 semesters, you could probably even get an MS degree in cs, for free for you as you should have a tuition waiver!

Astro_Dude

NSA told me to piss off. Several times. :(

Well, my tuition waiver is going to run out, and our group has no more money, so I'd be hard pressed to stay. More to the point though, everything is already in and I'm not even sure I can rescind the paperwork at this point.

twofish-quant

For companies in industry, the only way is to send in your resume and ask in the interview. One of the weird things is that industrial companies are often more tight lipped about what they are working on than government agencies, even those in defense or intelligence.

It would be nice to respond with "Great!!!! I'm ready to go!!!!"

There's a tricky balance between humility and arrogance that one needs to get right in the job interview. On the one hand, one doesn't want to come across as a know-it-all. On the other hand, one reason that physics Ph.D. training is good is that you can learn something new on your own very quickly. One good/bad thing about the modern work place is that employers will not spend money on training.

In finance, there are no long term projects. The longest term project that I know of has a deadline of two years from now. There are some vague long term goals, but no project plans that last more than a year, because things change too quick. Again that works for people with strong math and science backgrounds, since people are forced to learn things very quickly.

twofish-quant

Or you can take surf the internet and figure out how to program computers on your own.

StatGuy2000

I find it interesting that no one on this thread had mentioned anything about setting up a profile on LinkedIn, where you can upload a complete resume and tailor both your specific expertise and your career interests.

I would think that recruiters, headhunters and even a few hiring managers would contact the said individual, particularly if you have specific marketable skills, such as programming experience in MATLAB or C++ or statistical experience. I personally have received numerous contacts from recruiters informing me of positions (since I already have a full-time job which I enjoy, I have to turn these opportunities down, but I have connected with the recruiters to keep in me in the loop on other new opportunities).

StatGuy2000

Twofish-quant, I have a few questions based on your posts here:

(1) I find it interesting about your assertion to the uselessness of HR in financial firms in general, and more specifically in your firm. After all, part of the responsibility of HR is in ensuring that they recruit talent suitable for a given division or department. Shouldn't senior management in the quantitative division of the financial firm be providing more input and guidance to HR as to what general qualifications they are looking for (e.g. a doctorate or equivalent experience in math/physics, extensive programming knowledge, etc.)?

(2) I get the impression that financial firms that hire physics PhDs seem to be quite passive in terms of recruitment, in essence waiting for applicants to spam them with resumes highlighting their doctoral research. Am I mistaken about this? (in other industries it is quite common for potential employers to visit college campuses to do meet-and-greets with students).

(3) I am curious as to whether someone with a physics PhD has an edge in financial firms as opposed to someone with a PhD in math, statistics, or industrial engineering/operations research.

twofish-quant

Don't misinterpret what I say. HR is extremely useful to the firm. They are however, useless to you.

The reason is that 99% of the positions that HR fills are non-geek positions for non-geek applicants, so that filling geek positions is something that HR doesn't put a high priority on. Also, the research people would rather minimize HR's role in hiring. HR people tend to be people with *zero* technical skill, so when it comes to hiring technical people, both HR and the technical groups would rather they not make any actual hiring decisions.

What happens is in my firm is that we have internal people that do HR things. Often this is informal, (i.e. manager has a stack of resumes and asks for volunteers to go through them). If you are lucky, and you send your resume to HR then it will get forwarded to someone that can do something about it. However, there are a dozen ways for your resume to get lost in the shuffle, and there is no incentive at all for the company to make sure your resume doesn't get lost.

That's actually not the responsibility of HR when it comes to technical positions. We don't want them to recruit people for us, which is fine since it is something they would rather not do.

Doesn't work. Remember that people in HR do not have technical skills and they don't really have the desire to learn physics. So if you tell them look for people with deep knowledge of non-linear optimization, their reaction is "what's non-linear optimization?" The most you can do is to say "if anyone submits a resume with the keyword X, please send it to e-mail Y." And even that doesn't work well, because HR has no incentive to actually do that. Remember that everyone in the company gets hundreds of e-mails each day, and it takes time to sort through the e-mail, remember where to send it, etc. And why bother? The company doesn't care that you get the job. As long as the job gets filled by someone, everyone is happy.

Technical positions are different from other positions. For some positions, you don't need specialized knowledge to weed out incompetent people. For example, if I were interviewing for bank tellers or mortgage salesman, I can do that without too much specialized knowledge. (Quick: try to sell me a mortgage.) For other positions there is some credential you can use. For example, medical and law have licensing. For software engineering and quantitative finance, this isn't true. If you have no specialized knowledge and you are trying to interview a mortgage salesman, you can try to have them sell you a mortgage.

If you have no specialized knowledge, then how you are going to begin to do a technical interview of a Ph.D. in astrophysics? (Incidentally, I knew that I would like finance when people started asking questions about general relativity at the interview, and I knew I was in the right place when the interviewer pointed out that I flipped two variables in the answer.)

This is where headhunters come in.

We do meet and greets and "super-days". However the problem is the way that physics Ph.D.'s are distributed. There are only a few schools that graduate more than a dozen physics Ph.D.'s each year, and most schools graduate one or two each year. We do put on a show at the Ph.D. factories like MIT, but most physics Ph.D.'s don't get their degree from a Ph.D. factory.

It's not that we think that MIT physics Ph.D's are smarter than physics Ph.D.'s from Random State. The issue is that MIT graduates about 40 physics Ph.D.'s per year, whereas Random State graduates two, and so if you hold a "Ph.D. party" at MIT you can get people to show up. Also alumni are pretty key. A lot of these "meet and greats" are just an alumni showing up at their old school to give a talk.

Also, Ph.D. hiring is different from MBA hiring. MBA's get hired by the truckload. With some very rare exceptions, Ph.D.'s get hired one or two at a time. So if you are hiring one person, and you have one person graduating, then having a meet and greet isn't that efficient. Also, you have similar problems with the structure of finance. There are a few largest investment banks, but lots of mom-and-pop hedge funds. If you have a company that has twenty people, then it's hard to put on a road show.

The other problem is that campus career services is usually not set up for Ph.D. hiring.

In general no. However it can hurt or help you in specific situations, people do care a lot about diversity, so if you have a team of five people, and if three of them have physics Ph.D.'s, there is a very strong effort to make sure that the next person that gets hired isn't a physics Ph.D. On the other hand, if you have five people on a team and none of them have physics Ph.D.'s, then there is a bias toward adding someone that has done physics. Again, from the point of view of the job seeker, this looks random.

This is also why hiring decisions are team based. One thing that happens is that people tend to be biased toward people with the same backgrounds as they come from. I've noticed for example, that if you have someone that has done astrophysics, the I tend to give them higher ratings than if I interview someone with a background in statistics or operations research. Partly this is because, if I ask a question, it's likely to be something that someone that has done astrophysics knows.

But this can be bad, so what we try to do is to make sure that the interviewers have a mix of backgrounds so that the biases cancel each other out.

twofish-quant

It's not useful for finance. One reason it's not is that we've gotten specific instructions from HR, legal, and compliance about what we can and can't put in our linkedin pages, and I'm pretty sure that there is someone from HR that spends their day searching linkedin to make sure that policy is enforced.

But you are in the doghouse, if your own manager reads your linkedin page, and gets the impression that you are looking for work elsewhere. Something about finance is that much of the compensation gets paid as a lump sum at the end of the year, and so if your managers get the idea that you want to leave, that impacts your bonus.

Extremely dangerous to do in finance. Companies can't do this directly. If someone from company A looks on linkedin for people to hire from company B, then you are looking at a massive lawsuit. If you get lucky and happens to hit someone in company B, that wants to leave, then great!!! However, if you hit someone that is happy at company B, and hates company A because of something that happened years ago, you've just stepped on a landmine. That person will report the incident to their lawyers, their lawyers will talk to your lawyers, and you will be in big, big trouble.

Also this is why a lot of hiring in finance is personal. If I take a friend out for lunch, I can ask if he is happy or not (and tell him if I'm happy or not). I don't know what is going to come out of the conversation, but I know that he is not going to get me fired. If I talk to someone random, I don't know that.

Headhunters do this from time to time. However, in finance there are a lot of scummy headhunters, so one thing that you have to worry about is getting good quality HH's. One thing that I worry about is that some idiot HH will call me at work or send me e-mail at my work address. Now any non-idiot HH will know not to do that, but if I put my resume online, I have no way of screening out idiot HH's.

But even if I wanted to do this I can't, because if I did put my entire resume online, I'd get this nasty letter from HR telling me to take it down or I'll get fired.

Astro_Dude

Thanks again to everyone in this thread, my incessant whining must grow tiresome. :-p

I have started to do some basic C++ using just emacs and my g++ compiler I need for a variety of my NASA-made packages I run. I'm picking it up rather easy, but I'm having flashbacks to IDL with all this annoying declaring of variables. I much prefer Python.

I have been on LinkedIn for about 8 months and have not once been contacted by anyone for even a crappy job. Maybe that says something about my resume, because LinkedIn profiles are essentially resumes. I dunno.

I'm still having the issue that every time I look at even small DCs, I can't really find too much in the way of jobs I'm a great fit for... I just am not terribly optimistic applying for a job in engineering stuff (even when I can do it), because they just don't seem to care about physics people.

The REALLY frustrating thing is I have some contacts that are heads of entire divisions of their company, and generally the response I get back is that I have impressive credentials, they just don't have anything for me.

StatGuy2000

Twofish-quant, thank you for your replies to my questions, although your answers raise a number of other questions, as follows:

(1) As you mentioned, HR for the most part do not have technical skills and they are primarily responsible for filling the 99% of positions that are non-technical. However, many large companies (many financial firms in the US fall in this category) have large HR departments. You would think that within the recruitment division of HR you could find at least a handful of people with some limited technical capabilities who could devote themselves solely for helping to recruit technical positions.

For example, large pharmaceutical firms hire biostatisticians like myself, and much of that recruitment is handled through recruiters or HR (even though neither the recruiters nor the HR personnel have any background in statistics). Also, I have seen technical positions being listed on the websites of large financial firms in Toronto (where I live and work) which require a PhD in math/applied math/physics. After all, the role of HR is to ensure that the firm can fill a given position with the best possible candidate, and if the positions requires specific qualifications, such as a physics PhD, you would think that HR should be able to find candidates without necessarily needing to know "what" these candiates can specifically do (the interviewers can handle this aspect of things).

(2) I find it really curious that those who work in financial firms could get in trouble for putting your resume online in LinkedIn. I work for a clinical research organization (CRO) i.e. a contracting firm specializing in providing biostatistical services, among others, to pharmaceutical and biotechnology companies, and it is common practice for people in my general industry, from entry level personnel to senior management, to put their entire profile, including their complete resume, online. I have never heard of any of these individuals being reprimanded by HR or from management, simply for putting their resumes online.

(3) One thing I hear of constantly is the importance of networking, but it seems to be that financial firms seems to go out of their way to make this very difficult. Furthermore, financial firms, at least from your replies above, also seem to go out of their way to make it difficult for individuals with technical backgrounds (such as a PhD in physics) to know what positions are out there. Which leads me to wonder this, how are graduate students in physics or math interested in working for finance supposed to know who to apply to or where to send their resumes or CVs (including knowing which headhunter is reputable)? You had suggested simply spamming resumes left and right, but that's difficult to do if you don't even know which companies are offering what positions.

(4) On a related note to question (3), what does YOUR firm do to find suitable candidates for technical positions?

StatGuy2000

My suggestion is to have someone review your resume to see if you are highlighting skills that are especially marketable (example, statistical analysis or programming). Also, don't assume that people are automatically going to contact you just because you have a LinkedIn profile. It can take a while for someone to notice you, and I have also found, at least in my experience, that employers have contacted me during specific times of the year (typically during the months of January-May, and also around September-November).

I would also suggest trying to make either new connections on LinkedIn, send out messages to specific managers or recruiters, or join specific groups on LinkedIn that are focused on the area of your interest. You can find much information on potential job postings in this manner as well.