I did something like that once, many years ago. I was working on a doomed project to replace the numerical control for the companies main product line--internal grinding machine tools. The project was going no where. I was getting sent out of state to attend worthless meetings.
I took myself off the project, sat down and wrote an OS for the existing hardware and dragged a few coworkers along for the ride. By the time I left, almost all of the software group was working on my OS.
This was many years ago--I would never write an embedded OS today but this was the early days of microprocessors and good embedded OS software was very expensive and generally licensed on a per use basis. Trying to get the company to pay license fees would have killed the project immediately.
The OS and subsequent application software took about 2 years and worked quite well. It helped the company get a contract from GM Saturn. I learned a huge amount in those 2 years. It didn't help me keep my job at the company though. The broken nature of the company wasn't fixed by simply delivering working software. There was really nothing I could have done to extend my career there and in hindsight, I should probably have gone and done something different.
I took myself off the project, sat down and wrote an OS for the existing hardware and dragged a few coworkers along for the ride. By the time I left, almost all of the software group was working on my OS.
This was many years ago--I would never write an embedded OS today but this was the early days of microprocessors and good embedded OS software was very expensive and generally licensed on a per use basis. Trying to get the company to pay license fees would have killed the project immediately.
The OS and subsequent application software took about 2 years and worked quite well. It helped the company get a contract from GM Saturn. I learned a huge amount in those 2 years. It didn't help me keep my job at the company though. The broken nature of the company wasn't fixed by simply delivering working software. There was really nothing I could have done to extend my career there and in hindsight, I should probably have gone and done something different.