Start a two-person consulting practice, making sure to bill corp-to-corp and, if you can, avoid staff-aug projects that will simply stick you in one place for months at a time. Two people. Solo consultants get screwed by clients, have less time to manage and grow the business, and aren't learning how to operate as a company.
You'll learn how to market yourself, how to manage a sales pipeline, how to keep a set of books, how to track utilization and how to set up triggers for hiring new people.
If you're successful, you'll build a revenue stream that you can divert from to build a full-time product development team, at which point you'll have bootstrapped yourself and be at cash-flow-positive.
The ideas will come.
Working for a startup is fun. It will definitely teach you how to develop under hair-on-fire circumstances, and probably expose you to current technology (Rails instead of J2EE, Erlang instead of C++, etc).
But I was --- am --- shocked by how much 10+ years of dev work for startups did not prepare me to start a company from scratch. I got very lucky picking my partners; had I been even a little less lucky, I'd have failed in the first six months.
Respectfully, this doesn't make any sense to me at all. Being beholden to a BigCo gives you more opportunity to develop ideas than having control over a sales and delivery pipeline?
So, no. But I think you have a kernel of a valid caveat there, which is that as long as you're bringing money in that isn't coming from your product, you're going to be distracted. My proposed solution for that --- one that's working well for us --- is to get to a point where you can simply hire people to do one or the other. Hard to argue that a full-time product person is "distracted" because his paycheck is coming from a consulting practice instead of a VC fund.
This is also why I recommended starting a two-person company, and not simply becoming a freelancer. Freelancers do get sucked into "temporary W2" consulting hell.
You'll learn how to market yourself, how to manage a sales pipeline, how to keep a set of books, how to track utilization and how to set up triggers for hiring new people.
If you're successful, you'll build a revenue stream that you can divert from to build a full-time product development team, at which point you'll have bootstrapped yourself and be at cash-flow-positive.
The ideas will come.
Working for a startup is fun. It will definitely teach you how to develop under hair-on-fire circumstances, and probably expose you to current technology (Rails instead of J2EE, Erlang instead of C++, etc).
But I was --- am --- shocked by how much 10+ years of dev work for startups did not prepare me to start a company from scratch. I got very lucky picking my partners; had I been even a little less lucky, I'd have failed in the first six months.