The first company I worked at during the "dot-com" era, back when we were all worried about Y2K, lol, used genetic algorithms to solve mobile resource management problems - hence I developed an interest in "deep-search" problems, those problems whose solution space is too large to be solved precisely but solutions can be approximated via heuristics. I still have this interest, but have sort of found myself dealing with logistics types of solutions, i.e., working with solutions that "talk to" SAP supply chain solutions, fine, but does not sort of have the theoretical aspect which I have an interest in.<p>I am proficient with C and Java mostly along with UNIX commands and some auxiliary skills, but really have more interest in cool problems to be solved rather than the latest/greatest web fads like Ruby and so on. My question would be, how does one, having "cut one's teeth" so to speak in commercial software, sort of "transition" to more research. I also have a bit of a side-interest in AI as I have some experience with back-propagation neural networks written in C as a solution to predict energy markets. I think my biggest interest is sort of "big problems" that require heuristics to solve them but since while I think I have a pretty solid resume in terms of work experience, both in the area of mobile research optimization and in logistics, I don't have the academic "cred" (e.g. B.S. degree) to sort of "back that up" so to speak.<p>Most recently I have come across the concept of functional programming, of mathematical interest to me, so my current idea is to study a bit more about that. I just wondered if anyone in the HN community had any thoughts in terms of how to "transition" from a rather sucessful career in the commercial/enterprise software world towards more getting into the research/theory end of things, without perhaps the letters after my name to back that up, lol. Any thoughts appreciated! :-)
Apply to CS grad schools. See if you can work with any researchers or professors that will lead to publishable work. Read through their papers so you can see what they are working on and try contact a few people that have the same interests you do. (You mentioned genetic algorithms, neural networks and functional programming.) I've trying to do the same thing at the moment. Contact info is in my profile if you have questions. Good luck!
