As someone who has done this a few times (worked for large enterprise / found a problem worth to solve / started a company solving that problem / being acquired

As someone who has done this a few times (worked for large enterprise / found a problem worth to solve / started a company solving that problem / being acquired by other enterprise that wanted our solution exclusively): don't look for "ideas", but explore "problems". Working in an enterprise is great, because you are exposed - every day - to things that do not work in enterprise. For each problem you meet, ask yourself "would someone pay enough $$$ to get this problem solved once and for all?". If the answer is yes (and the problem is inspiring enough for you to dedicate 5+ years on that) the next phase is: what the distribution channel for a solution would be? Is the market too fragmented? How I would reach out to customers? What would be the cost to attract/retain customers? If you figure it out (and it seems a problem/distribution channel that works) - the next step is to think about a generic solution. Now you can test the "let's try to attract customers, for this problem, on my prototype/landing page". If the cycle starts to work, now it's time to figure out whether you want to quit and build the company/product/solution for real. The good of this process is that is 100% compatible with whatever NDA/NC/IP you might have signed, because you haven't written any lines of code up to this point. And when you leave, you have a solid framework of what you'll be doing next.