Where should the theory guys go get programming experience?

CloudyTrees

I'm graduating within a phd in theoretical cosmology in several months and have found that it is very difficult to land a job in the industry without any demonstrated programming experience. Knowing that the academic road isn't for me practically, I've read books on C++ and coded a little, but this, I guess, would hardly increase any probability with employment since it is not a "professional experience".

Where, then, could we theory guys gain any of this "professional programming experience" considering that no institution is providing any certification for C++?

We really appreciate any feedback!

mfb

Can you write some cosmological simulations? ;)
It is better than nothing. Depending on the job, it might be sufficient - and if not, it is something you can learn.

University courses in programming are another option.

ModusPwnd

My thought is taking some courses as well.

Here is an example of an online masters you could do in a year and then have a computer science degree. http://ecampus.oregonstate.edu/onlin...puter-science/

Locrian

I often see suggestions that one get involved in open source projects that you can later say you contributed to. I don't know how well this works, but you might ask around.

CloudyTrees

Thank you for the suggestions. I've used COSMOMC for two months (it's written in Fortran) but forgot most of it. Anyway I would definitely go back and review the material.

Thanks!

TomServo

I thought all theorists these days did loads of programming.

Locrian

I had three roommates in grad school, all three of whom were theorists, and none of whom ever did any programming. So it can happen.

Maybe things have changed in the past five years. One can hope.

CloudyTrees

As least our group doesn't. The only people that are using programs are all using Mathematica for checking analytical results.

MathWarrior

A lot of places if you have been programming in a language for many years its probably just as good as professional experience. The hard part is being motivated and finding things to do that test your knowledge. Which is often probably why they ask for professional experience, they want to see if you can follow goals and are driven to finish.

What do you mean by coded a little? Perhaps you need to be checking around for exercises online, such as project euler or even coming up with them on your own. I would think with an understanding of theoretical cosmology you could come up with some elaborate things to make?

If not, I have a suggestion try writing some cosmological simulators, or even (gasp) games. Either one can be pretty fun to do and give you some goals and experience.

SophusLies

That's pretty sad. Being a theorist myself, I have to say that programming is the most marketable skill a theorist can have. I guess if someone can land a professorship or postdoc within the same field of study (pen/paper theory) then they might never have to learn programming but the chances of that is pretty low.

Just out of curiosity, what did your roommates end up doing?

Locrian

Sorry, I said roommates, but they were officemates. There were more than a few days where the difference is blurry :)

One is a postdoc 5 years later. One ended up taking a job around the university that is probably minimum wage-ish in pay but does give him a little time for more chalkboard physics. One disappeared and I haven't been able to keep track of him.

I don't mean to make it sound like theory is a kiss of death or something. Those three guys are very bright and I'm sure they'll do fine. But most of us would like better opportunities than they had after graduation.

I know they did.

CloudyTrees

Yes you are correct, it was difficult for someone like me to get motivated in coding, as part of me used to think that using a computer isn't as cool as calculating by hand. And even today, I prefer to go paper/pen against computer as some kind of habit. I guess some theory students share this thought.

My coding experience is basically the COSMOMC project (using Fortran) which I almost forgot completely except the basic idea, plus some MATHEMATICA "coding" to produce some figures. But thank you for the projectEuler suggestion!

MathWarrior

If that is the case I highly recommend you pick up and learn C++ or C# in depth. I can suggest the following books either of which is pretty good for C++:

C++ Primer Plus (pretty good general book)
Programming: Principles and Practice Using C++ (bound to be a bit harder since its written by the guy who made C++, and I hear its full of lots of exercises)
C++ Without Fear: A Beginner's Guide That Makes You Feel Smart (pretty good for beginners maybe not what your looking for)

Knowing a language like C++ or even C# will dramatically increase your hiring potential at places other then just cosmological and physics related.

Even some of my professors with degrees in physics/math talk about how they wish they would of learned C++ long ago. I really found games to be the best motivator, they tend to have sufficient amounts of math (physics, optics, probability for AI, linear algebra, etc.) And they are fun to write, I don't know if you enjoy that but it might be fun for you.

If your looking for an easier route, I would learn C# its bound to have the same general performance. The reason they likely used fortran in your schools is because its pretty well known for having fast numerical computation abilities.

TomServo

Game programming is my backup plan. :)

Lavabug

I think video game programming is even more fiercely competitive than many science disciplines... gathering from what I know about people who went down that path.

Locrian

Agree with Lavabug. Make sure to have a backup plan for your backup plan!

cgk

OP, I think you might be gravely mistaken in your judgement about programming skills. These are practical skills, not something you just pick up a few months if you have never done this seriously before. Becoming a decent programmer takes at least a few *years* of doing this several hours per day! What you are looking for is equivalent to deciding to become a music producer because, well, a keyboard has only 88 keys and it can't be that hard to learn all of them. Or deciding to become a soccer star, because the rules are really quite simple.

You might want to have look into alternative fields of employment more suited to your skillset, or, alternatively, be prepared to go through some serious pain if you actually happen to land a job in programming.