Pike - Solving the "should we stop here or gamble on the next exit" problem

This is a problem I've been wanting to solve for about 5 years now. I found the rundown of the different approaches he tried very interesting.

I tried to prototype it once but gave up trying to find a good source for map and POI data. Pre-computing everything is a great idea... but I don't think it'd have occurred to me to spin up a giant AWS instance. I'd probably have gone into a rathole trying to break the problem into smaller chunks.