Job hunting advice for theoretical physics PhD

[NegativeDept]

I've often heard on PhysicsForums that career prospects are bright for theorists who paid attention to their math fundamentals and learned to program. That's me! It's also several of my fellow grad students. I suspect many others on this forum are in similar situations, and they may also benefit from advice.

Nobody is calling us. So if employers aren't finding us, we'll look for them. The question is:

What are some good ways to get our resumes, publications, portfolios, and/or selves in front of people who might hire us?

For the sake of generality, please assume the following conditions. These are all true for me, and most are true for my colleagues.

• We recently defended our PhDs or will do so within a few months.
• Academic, government, and private jobs are all open for consideration. We are flexible about location. (Exception: No working for hostile governments. We will not help North Korea decrypt your email.)
• We are competent programmers but lack the experience required for senior-level software jobs.
• We do not care if the work is "boring." Blue-sky basic research is awesome, but plain old statistics is also perfectly good.
• U.S. News did not rank our department "top tier," so we are not eligible for positions requiring degrees from high-status universities.
• We have applied to postdocs but cannot afford to bet all-in on the classic academic postdoc/professor/tenure trajectory.
• We do not live with our parents and therefore cannot afford to be adjunct instructors forever. (Adjuncts: Forgive my snark, but I think you know what I mean.)

I've partitioned job-search methods into subsets. Pardon the violent analogies; we do not intend to actually shoot anyone. If I forgot anything, please say so.

1. Buddy method: Ask friends/colleagues/family to introduce us to potential employers. Alas, this method is limited by the people we happen to know already.
2. Sniper method: Email people who are experts at stuff we like. I've heard that it's better to ask for advice about a specific topic rather than say "hey, gimme a job."
3. Shotgun method: Search for job postings and send a huge number of resumes and/or cover letters.
4. Headhunter method: Email recruiters who specialize in industries related to our research and training.
5. EDIT: Conference method: Go to professional and academic conferences. Talk shop with people and hand out business cards.

I have used 1,2, and 3 with no results so far. I did get a few polite "Thank you, but your background is not an ideal fit for us" emails. This is odd because what those companies do is damn near isomorphic to what I do. But that phrasing may be a euphemism for "Lawyers advise us not to say exactly why we rejected you." I have not used #4 because I don't have reliable info about which recruiters are shady and which are mostly honest. If you know any good ones, please promote them here!

In my case, I specialize in linear algebra, applied statistics, ordinary and stochastic differential equations, and a little cryptography and information theory. I do numerical simulations of all these things with NumPy, MATLAB, and Mathematica. I know more than enough C++ to code FizzBuzz, but not enough to impress a professional. I've been aiming for financial modeling, data mining, or quantum computing - but other suggestions are welcome.

My colleagues' abilities differ, but they're all good at basic probability, numerical programming, and either PDEs or ODEs. Most of them are also good teachers and skilled at explaining complicated things simply without dumbing them down or insulting the audience.

Any practical advice is greatly appreciated! (Ideological nonsense will be quietly ignored.)

 

[Vanadium 50]

Let's start with the simple questions. If Company XXX hires you, how exactly is that going to help their bottom line?

 

[Choppy]

I'll just throw a couple ideas out there...

1. Have you considered or attended professional conferences? I don't know who this would be for the markets you're talking about (South by Southwest maybe?), but if you wanted to get into, for example, medical imaging companies you may want to try attending RSNA's annual meeting, AAPM, or ASTRO. This unfortunately requires travel and conference registration, but in my experience these are great opportunities for face to face networking.

2. Job shadows? Rather than asking for an immediate position, you may want to just try to arrange some job shadowing days. This can allow you to meet people outside your immediate network and talk with people doing the work you're interested in. Even if that particular company is not hiring, people in the field are usually aware of who is hiring or at least what the market is like.

 

[NegativeDept]

Let's start with the simple questions. If Company XXX hires you, how exactly is that going to help their bottom line?

I suppose that depends on the specific industry. For example, suppose XXX is a hedge fund with lots of time-series data on historical securities prices $S_t$. I could write a program to find the securities whose $\log(S_t)$ has more large jumps than are predicted by a model (e.g. Black-Scholes, SABR) given its previous volatility. Then I could write a report warning traders that those securities are probably more risky than the models indicate.

Or maybe XXXCorp makes widgets, and demand for widgets is known to be stronger in the winter. I could deconvolve the widget demand history to produce a temperature-adjusted plot of widget sales so the company can better estimate how much of their recent gains/losses are just natural seasonal variation.

If teamed with a database expert (who need not be a statistician), I could do either of these things quickly for a large variety of different securities/widgets.

 

[NegativeDept]

1. Have you considered or attended professional conferences?

I was going to include that as "Conference method," but I totally forgot. I'm hoping to do that soon - though as you mentioned, it's tough when you're short on time and money.

2. Job shadows? Rather than asking for an immediate position, you may want to just try to arrange some job shadowing days.

That sounds like a very good idea. Do you know any good methods for arranging a job-shadow? My first guess is to cold-email companies, but I'm not sure who exactly to email or what the etiquette is.

 

[Vanadium 50]

OK, now what you need to do is focus on what you can do for the company - how are you going to make them money. That;'s what they are interested in.

 

[ParticleGrl]

So I tried variants of all of the above with fairly limited success. Eventually I was given the following advice, which did work
1. If you have done work that isn't directly industry relevant, give up on engineering/physics related work- there are far too many "ready-hires" with directly relevant backgrounds and you can't beat them out. (maybe the small improvement in the economy have made this easier).

2. find an in-demand field not opposed to physicists and sit down and learn the relevant field, and then start applying.

So while I was working as a bartender, I worked through some statistics and data-mining relevant stuff until I felt I was competent and then started applying. In the end, though, a fortuitous bit of luck landed me my first job- I was talking with a customer, it turned out he was looking to hire for a data mining group at an insurance company. He gave me his card, and I called the next day and was hired.

So if you want to do finance work, work through a few derivatives textbooks, code up a small project of your own, and then start applying. If you want to do datamining/stats work, learn the methods and again, code up a small project of your own. Basically, realize no one will hire you because you know physics, so you better learn something useful.

 

[NegativeDept]

So if you want to do finance work, work through a few derivatives textbooks, code up a small project of your own, and then start applying. If you want to do datamining/stats work, learn the methods and again, code up a small project of your own. Basically, realize no one will hire you because you know physics, so you better learn something useful.

I thought exactly the same thing, so I did a few mini-projects. One of them is even datamining/stats, though I suspect it will fail to impress a data science professional. I designed two statistical tests of online poker shuffles, and an astro PhD candidate extracted the relevant data from my Full Tilt hand history files. It's here if you're curious.

I think I'll try to combine your advice with Choppy's. The numerical method in my thesis can be used for commercial applications like autopilot error-correction and pricing rainbow options. I want to release it open-source, but I don't know how to advertise it to people who will actually use it. If I can afford it, I might be able to present it at engineering and/or finance conferences.

So I tried variants of all of the above with fairly limited success... In the end, though, a fortuitous bit of luck landed me my first job- I was talking with a customer, it turned out he was looking to hire for a data mining group at an insurance company.

That's kind of terrifying: you did all of these methods, and none of them worked as well as bartending!

This is way off-topic, but I can't resist... when you make an Old-Fashioned for yourself, do you mash fruit in the bottom, or just toss in a lemon/orange peel? (I have a friend who's a neutrino hunter and her husband is a bartender. They skip the fruit, but I go either way.)

 

[Choppy]

That sounds like a very good idea. Do you know any good methods for arranging a job-shadow? My first guess is to cold-email companies, but I'm not sure who exactly to email or what the etiquette is.

As far as the job-shadowing thing goes, you might want to ask at your school's student placement services office. Arranging job shadows is generally one of their functions so they'll know who to talk to. Personally I would avoid trying to go through HR and try to see if you can email someone as close to the position that you're actually interested in getting. I think cold emailing is fine (I get them every once in a while), and I don't really know too many alternatives.

One rule of thumb I've discovered is that people like to talk about themselves. So even if a full day job shadow is out of the question, most people would be willing to meet for an hour or so over lunch or coffeee to talk about why they do and their experiences in the field.

 

[ParticleGrl]

when you make an Old-Fashioned for yourself, do you mash fruit in the bottom, or just toss in a lemon/orange peel?

Always a lemon peel, never mashed fruit. Are you mashing fruit in place of the sugar? But hardly anyone ever orders an Old-Fashioned, and my preference for drinking is just whiskey on the rocks. I think I made less than 20 in several years of bartending.

I designed two statistical tests of online poker shuffles, and an astro PhD candidate extracted the relevant data from my Full Tilt hand history files.

For the sort of work I do, I'd suggest grabbing the data set for one of the competitions on kaggle.com and putting together a predictive model (a neural net, a random forest,etc) of some kind. I think to land a job you need to be able to answer some general questions about the process of building a model (how many features would you use in a model with X data points?, if you had hundreds of potential features, how would you narrow them down? what sort of model/approach would you try and why? Where would you expect your model to break down?) If you've worked through even one model, you probably have more context to answer these.

 

[pi-r8]

I've often heard on PhysicsForums that career prospects are bright for theorists who paid attention to their math fundamentals and learned to program. That's me!

It always confuses me when people talk about programming as if it's some binary thing where you either know it or you don't. It's a huge range of skill levels. If you're highly skilled at it- and you can prove it- then you'll have no problem finding programming jobs. If you only know the very basics (ie "hello world", arithmetic loops, that kind of thing) then that's not useful at all. If you're in between then it just depends on what specifically you've learned and whatever companies are hiring for at the moment. But just saying "I know how to program" is like saying "I know physics- can I be a physicist?"

 

[Arsenic&Lace]

Because if you're doing programming for theoretical physics applications, you are likely a). at least of intermediate skill, and b). may have even developed specifically useful skills. This is my experience working in a comp biophysics lab.

 

[carlgrace]

Always a lemon peel, never mashed fruit. Are you mashing fruit in place of the sugar? But hardly anyone ever orders an Old-Fashioned, and my preference for drinking is just whiskey on the rocks. I think I made less than 20 in several years of bartending.

Bit of trivia: Major Howard Armstrong, inventer of the Superheterodyne receiver and wideband FM drank an Old Fashioned with dinner almost every night. (he was the type of engineer to wear the same clothes and eat the same meals every day)

 

[jesse73]

Because if you're doing programming for theoretical physics applications, you are likely a). at least of intermediate skill, and b). may have even developed specifically useful skills. This is my experience working in a comp biophysics lab.

What simulation code would be just a "for loop"?

In a research setting you tend to need to make adjustments ( a lot of prototyping) to your model or speed up your implementation so that you would have to go out of your way to just learn basic programming in the 5+ years that a PhD program takes.

 

[jk]

I work as a programmer and I can tell you that for generic business programming type of stuff, the market is heating up. The software field is faddish though so you have to have skills that are in demand now. Things like Python and Ruby are in now...a few years ago it was Java, .NET etc. But just knowing a programming language won't get you far...you need things like good OO methodology, design patterns...these things take time and are best studied within a project. If you have free time, consider actively contributing to an open source project.
And learn SQL

 

[NegativeDept]

As far as the job-shadowing thing goes, you might want to ask at your school's student placement services office.

I'm going to visit the career center soon. I'll be sure to ask them about that sort of thing.

Personally I would avoid trying to go through HR and try to see if you can email someone as close to the position that you're actually interested in getting. I think cold emailing is fine (I get them every once in a while), and I don't really know too many alternatives.

That's definitely the plan. If I can talk to one of the lead math/stats/programming people, they'll get a lot more information than is contained on a resume. IMO for many subjects, most experts need < 15 minutes to approximately figure out whether someone actually knows the stuff they mentioned on their application.

My cold-email method is to ask a short, legitimate question that requires the recipient's expertise but not much of their time. For example: Do you think rejection-sampling RNGs like Ziggurat are worth the extra coding work? I'm using Marsaglia-polar and it seems fine, but I thought I should ask a professional. It gives them an excuse to talk about themselves and/or their subject without seeming rude.

For most of the competent TAs, emails like that from students are more likely to be answered. But there's a classic signal-to-noise problem with cold emails because they so often get ignored. When they do, it's hard to tell whether it's bad luck or I'm doing it wrong.

 

[NegativeDept]

It always confuses me when people talk about programming as if it's some binary thing where you either know it or you don't.

More precisely, I meant "We are competent programmers [in some language(s)] but lack the experience required for senior-level software jobs."

Because if you're doing programming for theoretical physics applications, you are likely a). at least of intermediate skill, and b). may have even developed specifically useful skills.

That's an accurate description of most of the physicists I know. Some of the biophysics people are very skilled at CUDA multithreading, and I've released a state-of-the-art open-source linear ODE solver. But few of us have ever made a working GUI.

The software field is faddish though so you have to have skills that are in demand now. Things like Python and Ruby are in now...a few years ago it was Java, .NET etc... If you have free time, consider actively contributing to an open source project. And learn SQL.

If I'm unemployed for a few months, I might use a convex combination of this advice and what ParticleGrl said. I know enough math and Python that I might be able to contribute something useful to NumPy/SciPy if I have the time.

Do you have an idea how long it takes to get competent at using SQL? I was going to practice C++ and R instead. I already know some C++ and a ton of statistics, so I have a head-start on these languages.

 

[Locrian]

SQL is pretty trivial. I remember when I first started learning it I found the logic of nested queries a bit weird, but it wasn't difficult to get through. Even big table joins and such are pretty simple. Maybe there's some super-use for the language that I'm not familiar with, but I've been using it for over four years and I get the job done.

Some of it will depend on how much work you're needing your queries to do for you. Personally, I prefer to manipulate data myself; I bring back as much data as I can handle and that doesn't lend itself to complicated queries.

 

[ParticleGrl]

Some of it will depend on how much work you're needing your queries to do for you. Personally, I prefer to manipulate data myself; I bring back as much data as I can handle and that doesn't lend itself to complicated queries.

So in the sort of work I do, 'as much as I can handle' (in memory) is usually significantly smaller than the whole data set, so I have to do the heavy lifting in SQL. Even so, I came in with minimal experience and have never had any trouble accomplishing what I want. Occasionally, I probably have to take a few more looks at optimization than more experienced people, but I at least get to the finish.

 

[jk]

But few of us have ever made a working GUI.

You don't need to. With your skill set and training, you would be a more natural fit for server side or "backend" programming

If I'm unemployed for a few months, I might use a convex combination of this advice and what ParticleGrl said. I know enough math and Python that I might be able to contribute something useful to NumPy/SciPy if I have the time.

Do you have an idea how long it takes to get competent at using SQL? I was going to practice C++ and R instead. I already know some C++ and a ton of statistics, so I have a head-start on these languages.

Depends by what you mean by "competent". It will take you probably a few days to learn the syntax for standard sql and be able to do simple queries. But it can take years to be an expert database developer or dba. For most developers, though, I'd say you can be reasonably expert after six months of use.

 

[NegativeDept]

For the sort of work I do, I'd suggest grabbing the data set for one of the competitions on kaggle.com and putting together a predictive model (a neural net, a random forest,etc) of some kind.

I started doing a mini-project which is kind of like that, I think. It's basically a simulated general practitioner: it asks a bunch of questions, looks at a medical database, and uses Bayesian inference to guess what the problem is. It's not on Kaggle, but I pitched it to a medical start-up who might actually use it or something based on it.

I'll keep an eye on Kaggle, though. If I can find the time, I might even be able to help with their SciKit-Learn project because I've already done a big pile of linear algebra in NumPy.

SQL is pretty trivial... Some of it will depend on how much work you're needing your queries to do for you. Personally, I prefer to manipulate data myself; I bring back as much data as I can handle and that doesn't lend itself to complicated queries.

Depends by what you mean by "competent". It will take you probably a few days to learn the syntax for standard sql and be able to do simple queries. But it can take years to be an expert database developer or dba.

I just got one of those Teach Yourself SQL In 1 Femtosecond books. Based on my tiny experience so far, it seems consistent with what Locrian and jk said: a few days to do basic stuff, and a long time to be a Grandmaster.

I'm hoping to find work that's heavy on numerical algorithms and light on advanced comp-sci, so I might be OK just knowing basic relational-database stuff for now. When one has a titanium screwdriver, it's tempting to solve every problem with a screwdriver and neglect all the other tools.

Always a lemon peel, never mashed fruit. Are you mashing fruit in place of the sugar?

I stir a small teaspoon of brown sugar, Angostura, and Peychaud's in the glass with a little splash of water. Then I rip a real (not maraschino) cherry in half, toss out the pit, toss in an orange slice, and stab at the fruit with a spoon because I don't have a muddler. Or when I don't have any cherries, I do what you said. I think that's the old-fashioned Old-Fashioned.

 

[Diracula]

Someone off-topic but related question: Does anyone feel that part of the problem physics Ph.D's have in finding employment in industry outside of their field of expertise is that a lot of hiring managers feel threatened by Ph.D. level physicists? Not necessarily by their skill set alone, but more what the ability to acquire that skillset implies about their talent and intelligence.

 

[ParticleGrl]

I started doing a mini-project which is kind of like that, I think. It's basically a simulated general practitioner: it asks a bunch of questions, looks at a medical database, and uses Bayesian inference to guess what the problem is. It's not on Kaggle, but I pitched it to a medical start-up who might actually use it or something based on it.

That should work. Just be sure you can answer questions like 'why did you pick Bayesian inference over other models/approaches', and make sure you held-out some of the rows from the database and used them to validate so you could create some metrics on how well it performs.

I stir a small teaspoon of brown sugar, Angostura, and Peychaud's in the glass with a little splash of water. Then I rip a real (not maraschino) cherry in half, toss out the pit, toss in an orange slice, and stab at the fruit with a spoon because I don't have a muddler. Or when I don't have any cherries, I do what you said. I think that's the old-fashioned Old-Fashioned.

If you are going to the trouble of making cocktails, you should get a muddler, a cheap one is <$5, and you will be the envy of your friends.

Also, unlike the link (weird to see Rachel Maddow in that context), I generally wouldn't use any soda water, just a sugar cube and some bitters muddled together, then whiskey then a bit of lemon rind.

 

[jk]

Someone off-topic but related question: Does anyone feel that part of the problem physics Ph.D's have in finding employment in industry outside of their field of expertise is that a lot of hiring managers feel threatened by Ph.D. level physicists? Not necessarily by their skill set alone, but more what the ability to acquire that skillset implies about their talent and intelligence.

No.

My take is that the difficulty that PhD's have in finding jobs in industry has more to do with their lack of skills that are immediately useful to hiring managers.

Incidentally, although I don't have a PhD, I believe the acquisition of a PhD in any field says more about a person's capacity for hard work and perseverance than their talent or intelligence, however vague those terms are.

 

[TMFKAN64]

Someone off-topic but related question: Does anyone feel that part of the problem physics Ph.D's have in finding employment in industry outside of their field of expertise is that a lot of hiring managers feel threatened by Ph.D. level physicists? Not necessarily by their skill set alone, but more what the ability to acquire that skillset implies about their talent and intelligence.

No.

The real problem is coming in with this attitude when, in reality, you are probably underqualified for the position.

 

[Diracula]

No.

The real problem is coming in with this attitude when, in reality, you are probably underqualified for the position.

I don't have a Ph.D.

I have, however, come across some rather nasty attitudes towards people with physics/math/engineering backgrounds when they interact with scientists with non-quantitative backgrounds (think biology). I only have a B.S in physics, and there have been a LOT of positions that I've been pretty much optimally qualified for (as in, I don't think it would be possible to find someone else that matches their job requirements as closely as I did), and I wasn't even contacted for an interview. Luckily I found a supervisor that views physics positively, and the difference in how he treats me vs. co-workers (supervisors included) in previous jobs is unbelievable. I personally find it hard to believe that this attitude doesn't carry over when making hiring decisions, and I don't think you'll ever hear someone in a hiring meeting say "wtf this guy has a physics Ph.D. he's too smart for us to hire", so you kind of have to read between the lines sometimes.

Additionally, if a Ph.D. implies so much about work ethic (rather than intelligence), why is it so hard for people with intermediate or better programming skills to find a job? Shouldn't companies really really want to hire people with an excellent work ethic and solid programming skills who obviously have a capacity to learn difficult material quickly?

 

[TMFKAN64]

Additionally, if a Ph.D. implies so much about work ethic (rather than intelligence), why is it so hard for people with intermediate or better programming skills to find a job? Shouldn't companies really really want to hire people with an excellent work ethic and solid programming skills who obviously have a capacity to learn difficult material quickly?

If there is a problem having a Ph.D. and applying for a programming job, it is probably that the perception that the only reason the person is applying is that he or she is desperate for a job and will leave as soon as a physics position becomes available.

Being able to learn difficult material quickly is a plus, of course, but not as much a plus as already knowing the material. There is a big difference between writing a few thousand lines of simulation code and being able to debug someone else's code in a large collaborative project.

 

[jk]

I don't have a Ph.D.

I have, however, come across some rather nasty attitudes towards people with physics/math/engineering backgrounds when they interact with scientists with non-quantitative backgrounds (think biology). I only have a B.S in physics, and there have been a LOT of positions that I've been pretty much optimally qualified for (as in, I don't think it would be possible to find someone else that matches their job requirements as closely as I did), and I wasn't even contacted for an interview. Luckily I found a supervisor that views physics positively, and the difference in how he treats me vs. co-workers (supervisors included) in previous jobs is unbelievable. I personally find it hard to believe that this attitude doesn't carry over when making hiring decisions, and I don't think you'll ever hear someone in a hiring meeting say "wtf this guy has a physics Ph.D. he's too smart for us to hire", so you kind of have to read between the lines sometimes.

I haven't come across such attitudes. If anything, having a degree in any quantitative field seems to give people impressions about the capability of that person that are not always warranted. The reasons you were not called back for the interview can be many, including that there were people who applied and were more qualified than you.

Additionally, if a Ph.D. implies so much about work ethic (rather than intelligence), why is it so hard for people with intermediate or better programming skills to find a job? Shouldn't companies really really want to hire people with an excellent work ethic and solid programming skills who obviously have a capacity to learn difficult material quickly?

Because there are many more people with intermediate or better programming skills and excellent work ethic than there are jobs.

 

[Diracula]

I haven't come across such attitudes. If anything, having a degree in any quantitative field seems to give people impressions about the capability of that person that are not always warranted. The reasons you were not called back for the interview can be many, including that there were people who applied and were more qualified than you.

Like I said, unless the job ads were completely fabricated and had nothing to do with the actual job, it was really not possible to be "more qualified" unless the additional education in physics was viewed as a negative.

On at least one occasion I knew some of the candidates that received interviews. There was no way it was because they were more qualified. On one of those occasions I had insider information on the selection process. Trust me, sometimes people are threatened by people with backgrounds in fields like physics if it is not typical for those jobs.

I have no knowledge of the software industry though, and I can see how it would be different there because of the strong correlation between math ability, abstract reasoning skills, and programming. This isn't as necessary in the non-bioengineering fields of biology, and I think people with quantitative skills are not viewed in a positive light by some in those and similar fields. I could have just had terrible luck at my previous jobs, but that is what I observed and my hypothesis was pretty much confirmed in several ways.

Because there are many more people with intermediate or better programming skills and excellent work ethic than there are jobs.

I acknowledge that the attitude in programming may be very different. However, are there really so few jobs and so many outstanding candidates that people with at least an intermediate knowledge of programming and a Ph.D. in a field as difficult as physics can't find reasonable employment? I thought programming jobs were relatively abundant -- and I hear stories all the time about how novice (yet employed) programmers can't program a lick. It just doesn't quite add up to me that physics Ph.D.'s can't find jobs and it's only because there are all these other excellent unemployed job seekers out there who are more talented and more skilled than a Ph.D. level physicist.

 

[ModusPwnd]

Like I said, unless the job ads were completely fabricated and had nothing to do with the actual job, it was really not possible to be "more qualified" unless the additional education in physics was viewed as a negative.

More qualified means you have more experience or the right experience. Fresh outs are generally the least qualified, even in the upper levels of PhD holders.

 

[Diracula]

And sorry for the derail. I guess my point would be don't necessarily emphasize your physics background too much and don't get too abtract and technical when talking about your work in the application process if 1) the job doesn't call for that skill set, and 2) it would be over the hiring manager's head.

I acknowledge this is probably less of an issue in the programming industry. I guess it all boils down to don't let yourself be viewed as a threat by people with the power to hire you (or promote you should you get the job).

 

[Diracula]

More qualified means you have more experience or the right experience. Fresh outs are generally the least qualified, even in the upper levels of PhD holders.

I wasn't fresh out. Just relating my experience on how people with physics/math backgrounds are often times viewed by people without such a rigorous background in quantitative fields.

If you really think this phenomenon doesn't exist at all such that you can explicitly ignore it, I would suggest reading some books on social power dynamics. It's actually probably particularly useful to know this type of stuff in a bad economy where jobs are scarce.

 

[carlgrace]

Work your network and show initiative. That's the best advice for getting a job you'll ever get. A surprisingly large number of job listings aren't real in the sense that the organization already has a candidate in mind and they are just satisfying employment regulations and requirements. I have seen that so many times it boggles my mind.

There are exceptions, but the best way to get a job is to ask around. Ask people you know if their organization is hiring, or if they know anyone who is. Get introduced to people who can help you. In my job, I've never hired someone who just sent in a resume responding to a job listing. They either reached out through channels or I reached out to someone who impressed me at a conference or similar.

You can do it, but you need to get your advisor to pry open some doors for you.

 

[jk]

And sorry for the derail. I guess my point would be don't necessarily emphasize your physics background too much and don't get too abtract and technical when talking about your work in the application process if 1) the job doesn't call for that skill set, and 2) it would be over the hiring manager's head.

You should always emphasize those things in your background that are useful to the employer. Otherwise, why would they be interested in you?

I acknowledge this is probably less of an issue in the programming industry. I guess it all boils down to don't let yourself be viewed as a threat by people with the power to hire you (or promote you should you get the job).

I have never seen a situation where anyone was seen as a threat for being smart or capable. Are you sure you are drawing the right conclusion from your rejections?

 

[Diracula]

You should always emphasize those things in your background that are useful to the employer. Otherwise, why would they be interested in you?

As I specifically said in the reply you quoted: "the job does not call for that [specific] skill set". They would be interested in you if you convince them you can do the job they want you to do, and, additionally, they feel you won't cause any disruption in their ascent up the corporate ladder.

I have never seen a situation where anyone was seen as a threat for being smart or capable. Are you sure you are drawing the right conclusion from your rejections?

Then you are lucky. This is actually a well established phenomenom, and is relatively common.

I'm not sure that I'm drawing the right conclusion from my rejections because no one told me "we didn't hire you because we're threatened by you". It is possible the person I knew who was in the meetings where they discussed candidates for the job was lying to me. And it is possible that the fact I was systematically pulled off projects that I was extremely successful at happened to coincide right around the time I was incorporating ideas from fields close to physics in my work (and invited to give a presentation by another group for this work). It could be a further coincidence that I was specifically told that my physics ideas weren't welcome in my performance review (nevermind it was encouraged right up until I could demonstrate results). And it could also be a coincidence that all of this success coincided with my former research group beginning to repeatedly talk down to me and act like I was an idiot at every possible opportunity, and everyone else being blown away that I was being treated like this.

But yeah, I'm not completely sure I'm drawing the right conclusions. All of these things could be coincidences and my friend was lying to me to top it off. Like I said, if you don't think this ever happens, you believe people are perfectly rational and never try to outcompete you for scarce resources (promotions, power, money, status), then feel free to completely ignore the possibility that hiring managers ever view people with a demonstrated capacity to master theoretical physics as a threat.

Are there really so many amazing unemployed people out there that physics Ph.D.'s with good programming skills struggle to find a SINGLE relevant job and remain unemployed for months, sometimes years? I know a lot of people who successfully found technical jobs at the peak of the recession, and didn't have too much difficulty. They were not, I repeat were not, as capable, smart, or qualified as some of the physics Ph.D.s I know or that post on here. Why is that?