Now, for design and software, that works really well. But other things come against basic physics.
Mr. Musk must realize a productivity improvement by a factor of 10 if he doesn't want to be 'bored' with the whole thing. He can't get that with actually digging the tunnels. It costs so much to excavate a cubic meter of soil and that's that. In past 30 years, giant digging machines have improved things maybe a factor of two, but they always get stuck. We are now flatlining the economics.
While I'm writing this, I'm thinking he can get a factor of 10 in operating the tunnels, by using a giant packet routing system, but first you have to dig. And go through a zillion faults in LA. Good luck in getting the busgus to admit that they don't mean anything - their lives are on the line. :)