Weekend chat: What jobs give opportunities to do sys prog PLUS own the project?

The thought came from reading this HN post: https://news.ycombinator.com/item?id=29104047 I realized, while reading the post, that those programmers in JPM/BAML/etc. got the chance to do some really interesting system programming, including creating a whole new IDE, some hacky and brilliantly implemented heavy-weight libraries and frameworks. And the best thing is, they got adopted by thousands, if not tens of thousands of users without any marketing. Look at it this way. Let's say you got the chance to join the Visual Studio team, or Jetbrain, what is the probability that you are to work on AND own a brand new IDE? Probably ZERO. You might say, "But I can put up my own and open-source it", my answer is "You would be really lucky if three digits of users pick it up". After some thoughts, I guess here is my conclusion. The following prerequisites must be satisfied if one wants to land in such a position: - The company must be a "closed-source" company, i.e. open source can come in but the code doesn't necessarily get out; - The company must NOT have a ton of brilliant programmers, i.e. it increases the chance of someone, especially someone new to helm a new project. - Extended from previous one: The company is NOT a technology company but it has to be a rich one (a bank for example). So that once it adopts something new it is willing to stick to it, regardless of the cost (Companies such as Google and Microsoft are less willing to do so, just look at how many products they drop) - You must wait until the right moment to join (otherwise, if you joined earlier, you probably only got to maintain a shit mountain of COBOL code, and if you joined late, you probably only got to maintain a shit mountain of Python code) - It is imperative that your Director wants some change in the IT system, either of political reason, or technical reason. The reason is not important, and whether the new system benefits the company is not important either. The important thing is some big shot wants some change, and you happen to be the one picked. I guess this kind of one-time-in-life chance cannot be scientifically analyzed, but I think there might be a way to generalize the phenomenon and gives oneself a better career. What's your thoughts my friends?