As a developer, I know that I waste far too much time building my own software for things that are already well solved. At this point I'm famous (amongst my friends) for yak shaving, and while I personally enjoy it, sometimes it's just better to get shit done.
The only way you can avoid double work is knowing about a piece of software that does what you need.
The only way to know about all the awesome software out there is to:
I know I can't do any of those, so I started writing a newsletter that sends out a new (to you, hopefully) F/OSS project every day.
Here's the list so far (if you don't know about these projects, check them out):
So I guess I'm doing something kind of antithetical -- I'm suggesting you skip one of the biggest pitfalls of being a developer-founder. Don't develop things if you don't have to so that when you do, you can go all out.