Strategy for going from postdoc to software

[js54434]

I'm finishing up a physics postdoc in experimental condensed matter physics, and I'm looking to go into industry. I'm aware that one avenue is programming, but I would be competing for entry-level positions against people fresh out of college who are specifically trained in the field, while my main experience is with data analysis in python (using numpy and matplotlib to fit curves and generate plots, etc.) As a physicist with the typical data analysis skills, what kind of programming jobs would maximize my comparative advantage and chances of being employed?

In addition to Data Science, are there any specific types of programming jobs that I should be looking at, jobs which are both relatively numerous but also allow someone with a physics Ph.D. to have an advantage?

(I say "relatively numerous" because I see posts on Reddit recommending things like game development, but if there aren't that many jobs in that sector then it's not a good strategy to focus on it even if physics PhDs would have an advantage. I'm not limiting myself to the software industry and have applied for jobs in the semiconductor industry, defense industry (in the USA), a consulting position, and a data science position, thus far without luck. I haven't applied yet to software jobs because I'm at a loss as to where to start.)

 

[FactChecker]

You might want to expand a little on your "data analysis" background. How are you at statistics? Data analysis with statistics may have jobs to offer.

 

[js54434]

By "data analysis" I mean mostly curve fitting and generating plots. The most complicated thing I've done (to satisfy my own curiosity) was to calculate neutron scattering intensity of a system the hard way by doing a simplified molecular dynamics simulation, then computing the relevant correlation functions.

My statistics knowledge is pretty elementary. For instance, I know that the central limit theorem holds under many circumstances but not with Lorentzian distributions. However, I don't know what a "confusion matrix" is (to name one random thing I saw in a job posting), and I don't have experience with machine learning. I am applying to certain data science positions, and I'm sure I can learn whatever additional statistical knowledge I need to if I get the job, but I'm also curious about jobs in the software field because I've heard that it's a not uncommon avenue for physicists going into industry, and I'm trying to maximize my chance of staying employed.