I can't understand why the model of "build an outstanding App, then reduce engineer team to < 100" is not a thing. If they could do this, they'd be profitable.

I can't understand why the model of "build an outstanding App, then reduce engineer team to < 100" is not a thing. If they could do this, they'd be profitable. And I think they could do this, if their codebase wasn't a mess. Heck, I bet 25 excellent googlers could make a run a service like this with 99% uptime (obviously there would need to be a large support staff still).