I've supervised a couple of undergraduate projects. Parallel sorting algorithms are a cool topic but quite specialized, so it would seem unlikely that you could

I've supervised a couple of undergraduate projects. Parallel sorting algorithms are a cool topic but quite specialized, so it would seem unlikely that you could train yourself up to an employable standard within an academic year in addition to your other responsibilities. A better use of your time would be something to get you hired. I recommend that you organize your thesis as a comparative study of two or three competing technologies you'd be interested to learn, by implementing a similar application with each of them (e.g. Ruby vs. PHP, Java vs. Objective C, MongoDB vs. Cassandra, etc. etc.). Have at least a few thousand lines of working code if you want to be taken seriously by an employer, and throw in some critical thinking to please the academics.