It's a bit of a side point, but this is one reason I almost insist on incorporating some of what I do at work into open source projects. I realise there are places that make this very difficult. Sometimes you have to do it in reverse - find an existing open source project, get that in the door at work, then contribute to that project. Done that way it doesn't seem like the same giving away the corporate crown jewels as open sourcing existing things originated in-house.
But the result is you don't end up 25 years downstream with nothing to show for it. You end up with a collection of open source projects to your name that you can keep working on and re-deploying over your lifetime.
But the result is you don't end up 25 years downstream with nothing to show for it. You end up with a collection of open source projects to your name that you can keep working on and re-deploying over your lifetime.