Build a telescope. There are many links to plans, here's one: http://www.telescopemaking.org/
There are lots of opportunities for electronics with a telescope: motor mount, autoguider, goto integration, etc.

Americans might not have been laughing, but Icelanders sure as shit were!

I got my PhD last year in a similar field, and I made the leap into video game development. Best decision I ever made. Besides the fun factor, the biggest issue in a science career (well, in academia at least), is just how freakin long it takes to get your career going. After grad school, 6 years in my case, you have to do a post-doc for a few years, then get (hopefully) get a tenure-track faculty position, then work your ass off to actually get tenure. Then, if everything goes right, your career starts. For me, that time commitment was simply not worth it.