Transitioning from Theoretical Physics to Software Dev: Steps & Difficulties

[ATmundo]

I'm going into my final year of undergraduate theoretical physics. I have had a bit of a deviation in what I want to do and would really like to work in software development in the future. How difficult would it be for me to deviate from my current position and what steps should I take to to enter this field. Thank you very much.

 

[berkeman]

Welcome to PF.

What programming languages have you learned so far? What is motivating you to want to switch career plans? What kind of software do you want to develop?

 

[Vanadium 50]

And what country are you in? In what country do you want to work?

 

[ATmundo]

Welcome to PF.

What programming languages have you learned so far? What is motivating you to want to switch career plans? What kind of software do you want to develop?

I have only learned python and a dabbled a bit in C++. I don't necessarily want to change career I just had no exact career in mind when I started my degree. I did physics partially because I thought it would be useful but mainly because I enjoyed it and wanted to learn. I have no exact type of software that I want to code but ideally one which allows me to solve cool problems and produce something helpful and meaningful. I like software development as it allows for a lot of control over the type of life you want to live. I want to have a good work life balance, low stress and yet still make decent (not necessarily amazing) money. It also allows me to program which I find fun and solve problems which I really enjoy.

 

[ATmundo]

And what country are you in? In what country do you want to work?

I am from Ireland and would like to continue working in Ireland.

 

[berkeman]

How difficult would it be for me to deviate from my current position and what steps should I take to to enter this field.

So do you want to also earn a degree in CS, or are you asking what you can do informally outside of your Physics degree to make you a good candidate for software jobs?

The CS classes in university are pretty specialized and different from other Engineering classes (let alone Physics classes). It's important IMO to at least take classes in Data Structures and Object Oriented Programming and other CS specific classes to make you better at architecting and coding software projects. I also found my class on Compiler Design to be very interesting and valuable, since understanding how compilers work has helped me a lot in working on the more complex software projects that I've been involved with.

So if you want to also earn a CS degree, it would probably add at least a year or more to your university time. If you want to be considered for employment in some CS area, you would either need to learn all that stuff on your own (not easy), or choose a specific type of coding and the appropriate languages, and consider attending a "Coding Academy", which takes about 6 months of full time work, at least for those academies that we have here locally in Northern California.

 

[ATmundo]

So do you want to also earn a degree in CS, or are you asking what you can do informally outside of your Physics degree to make you a good candidate for software jobs?

The CS classes in university are pretty specialized and different from other Engineering classes (let alone Physics classes). It's important IMO to at least take classes in Data Structures and Object Oriented Programming and other CS specific classes to make you better at architecting and coding software projects. I also found my class on Compiler Design to be very interesting and valuable, since understanding how compilers work has helped me a lot in working on the more complex software projects that I've worked on.

So if you want to also earn a CS degree, it would probably add at least a year or more to your university time. If you want to be considered for employment in some CS area, you would either need to learn all that stuff on your own (not easy), or choose a specific type of coding and the appropriate languages, and consider attending a "Coding Academy", which takes about 6 months of full time work, at least for those academies that we have here locally in Northern California.

I wouldn't mind doing a masters program. Like ideally there would exist a 1 year or so masters that accepts theoretical physics students and gives them the necessary skills to get a software development job. Do such masters exist to your knowledge or would the things I need to know be too elementary to be contained within a masters program?

 

[berkeman]

I'm no expert on that question, and you may have to talk to your local universities to see what they can offer, but it's hard for me to believe that you can get into a Masters program in CS without having taken at least the 5-10 undergraduate prerequisite classes. Masters work is usually more specialized in nature (with different areas of specialization), with no time for re-taking basic CS classes.

 

[ATmundo]

I'm no expert on that question, and you may have to talk to your local universities to see what they can offer, but it's hard for me to believe that you can get into a Masters program in CS without having taken at least the 5-10 undergraduate prerequisite classes. Masters work is usually more specialized in nature (with different areas of specialization), with no time for re-taking basic CS classes.

Yeah that makes a lot of sense. I appreciate the response. I am meeting with a head of CS in one of my local universities and I'll ask him how he thinks I should approach this.

 

[pbuk]

I think the answers you have received so far are somewhat US biased.

Firstly undergraduate degrees in Europe don't generally work with the "mix and match" style of US degrees; there will be a couple of CS related courses on a Physics degree programme (usually computational stuff in C++ or Python which is what I assume you have done) but there's no way you could take OOP or compiler design.

But secondly, on the other side of the coin, it is very common for employers with big IT intakes to recruit science, and particularly physics, graduates.

So to answer the question in the title of the thread, I'd say its probably easier to "transition" from an undergraduate theoretical physics degree to a career in software development than it is to stay in theoretical physics.

 

[Hornbein]

Wall Street math is largely about differential equations, so physics degree is an advantage there.

If I were you I'd find the kind of job I wanted to do, then ask the people who do the hiring what they are looking for. It isn't a secret.

 

[ATmundo]

I think the answers you have received so far are somewhat US biased.

Firstly undergraduate degrees in Europe don't generally work with the "mix and match" style of US degrees; there will be a couple of CS related courses on a Physics degree programme (usually computational stuff in C++ or Python which is what I assume you have done) but there's no way you could take OOP or compiler design.

But secondly, on the other side of the coin, it is very common for employers with big IT intakes to recruit science, and particularly physics, graduates.

So to answer the question in the title of the thread, I'd say its probably easier to "transition" from an undergraduate theoretical physics degree to a career in software development than it is to stay in theoretical physics.

This is a hopeful reply. I appreciate the response.

 

[haushofer]

In Holland: absolutely no problemo. If you know Python and how a computer is turned on, lots of companies are interested.

Edit: what pbuk says.

 

[crashcat]

A lot of software companies hire for specific skills, so having a bachelor's in physics plus any sort of concrete demonstration in skill should get you a job. Class projects or mini research projects or something you did in your own time - highlight that in your resume and be well prepared to talk about it. Real projects not just "I took this class." You shouldn't have much difficulty transitioning if you're motivated to keep programming on your own time until you get hired.

Edit: I would avoid additional education if possible. Just get your first job. Once you have a B.S. Physics plus 2-4 years experience in software, it won't matter that you never did whatever 12 or 18 month software crash course you could possibly have done. Fortunately you still have a year left to build a real software portfolio if you feel it is lacking right now. Write a beta app even if you ultimately don't commercialize it, write physics simulations, dabble in DFT, make a silly game, join some sort of FOSS project and just read thousands of lines of code and make a few minor contributions when you feel confidant. Anything you do in the next year can help, but some are better than others.

 

[MathematicalPhysicist]

So I guess the money issue is a big concern to you. (For example of having kids, etc).

I Cannot blame you, if I weren't as obsessed in maths-physics-engineering-logic I would too may have opt to go to the industry. But the powers of be didn't let me...

 

[Eclair_de_XII]

If you know Python and how a computer is turned on, lots of companies are interested.

You make it sound easy to find work in Holland. From what I've seen while looking for work in the U.S., most companies want candidates with years and years of experience programming in various languages, years of experience coding in those languages in a professional environment supplemented by a college education, and numerous other qualifications that I do not usually bother to read or understand, because by the time I finish skimming the list of requirements for a given position, it is more than apparent that I am unfit for the role. That being said, as computer science was not my major, I cannot disagree.

 

[pbuk]

most companies want candidates with years and years of experience programming in various languages, years of experience coding in those languages in a professional environment supplemented by a college education, and numerous other qualifications

This is not relevant to this thread which is looking at graduate entry positions as the OP is

going into my final year of undergraduate theoretical physics.

 

[Psi Star]

I have a PhD in theoretical physics and, during the course of my career in applied physics, I have learned at least nineteen different computer languages. I design, build, and test prototype systems that are then passed to engineers for manufacture. For me, programming is a necessary tool but not an end in itself. I recommend that you stay in physics, acquire strong math skills, and consider programming as a means to an end rather than an end in itself. Your saleability will be much better.

 

[enricfemi]

How about apply physics? It needs both physics and CS.

 

[symbolipoint]

I like software development as it allows for a lot of control over the type of life you want to live. I want to have a good work life balance, low stress and yet still make decent (not necessarily amazing) money. It also allows me to program which I find fun and solve problems which I really enjoy.

You must not count on that. I am acquainted with a software developer, and he is forced to endure much stress. He has also suffered a couple of lay-offs.

 

[symbolipoint]

during the course of my career in applied physics, I have learned at least nineteen different computer languages.

How does or did that happen? How effectively? How so many?

 

[Hornbein]

How does or did that happen? How effectively? How so many?

Computer languages are limited in scope. They are all basically the same. It's matter of learning the niggling details and gradually picking up speed, which can take years. Learning the basics might take only a few hours.

The rules of logic are very simple. True false and or not. Program flow is if then else.

 

[symbolipoint]

Computer languages are limited in scope. They are all basically the same. It's matter of learning the niggling details and gradually picking up speed, which can take years. Learning the basics might take only a few hours.

The rules of logic are very simple. True false and or not. Program flow is if then else.

Thanks, since I guess you're right. That did not do me any good when I was struggling long ago. (As one would expect, my degree was not in Computer Science.)

 

[bryantcl]

I know I'm late to the game but I'll contribute my $0.02 anyway as it may help someone other than the original poster. I've worked in software development/IT for 25 years and I have interviewed many candidates. The first things I look for are attitude and initiative. These are things that cannot be taught. The third thing I look for technological ability. If you slogged through a theoretical physics curriculum I believe you also have aptitude. With these things the technology used in a particular job will not matter as much.

You said you were in Ireland so a postgraduate certificate or diploma or perhaps a coding bootcamp (I don't know how big those are in Europe). This may have shell out some money to get to the point you want to be but sometimes taking the step back to go forward again is worth it.

Best of luck,

clb