There are various theories along the lines of "principal engineer retention strategy", where part of the reason for these projects (and I have to emphasize "par

There are various theories along the lines of "principal engineer retention strategy", where part of the reason for these projects (and I have to emphasize "part" in "part of the reason") is because they are the pet projects of some particular individual at Google who can deliver tons of value to the company other ways, and who might leave if they don't get to work on their pet projects. I have definitely heard stories of people at companies like Google and Facebook who have done things in their career that saved or made the company enormous amounts of money in one fell swoop, people who can help get massive projects back on track or solve thorny resource usage problems. The theory is that if they've done it a couple times in the past, you do what it takes to keep them around because they'll probably do it again in the future. These people generally get positions crafted for them, and sometimes teams.