If I had to pick one piece of advice for software developers looking to jump into entrepreneurship, it is that most companies fail because they build something

If I had to pick one piece of advice for software developers looking to jump into entrepreneurship, it is that most companies fail because they build something no one wants. In fact they build it really well. Thus technical debt, scalability etc. simply don't matter until you iterate your way to solving a problem other people care about. That's much better than solving a problem well, a problem that not enough people have. Ie. in short, stop engineering software and start figuring out what people actually need. Not just 'nice to have', but a real need that causes real pain. To see if enough people need what you are planning to build - you don't need to built at all, just draw it out, explain it in a doc, and go ask people. You have to really ask them and push beyond their initial "sure yeah, it'd be nice to have". Ask them how they do the task / fulfill the need now. Ask them how much that costs. How much would they pay if you built a better one, etc. Really try to get a "no i don't actually need it" instead of being content with the polite lie of people wanting the product.