[Manhattan Project] I've been reading Rhodes's The Making of the Atomic Bomb—which is fantastic, by the way—and while he details the timeline going from Fermi's handmade pile at the University of Chicago to having Oak Ridge and Hanford up and running, I don't understand how it was possible to make the jump from "first-ever atomic pile" to "two fully-operational nuclear power plants" in such a short period of time.
How the heck does someone sit down and design the first full-blown nuclear power plant from scratch? Are there any books that address this question?
The short answer is, it was very difficult and they had to make a lot of educated guesses. And they didn't do it perfectly by a long shot. They took what understanding they had obtained from both CP-1 and the smaller pile experiments, which gave them an understanding of how the reaction would scale with size. Then they just scaled it up in almost one jump. (I say "almost one" because they built the X-10 reactor concurrently and used it for further pilot-scale work.) Which is not standard scientific or industrial practice, which is why DuPont wanted contracts that explicitly said it was not on them if this didn't work right.
One of the famous anecdotes (probably in Rhodes somewhere) is that the scientists gave the size of the piles and the engineers added a safety factor of some excess to the designs, which the scientists thought was silly. In the end it turned out to be necessary because of the xenon poisoning issue, which had not really been anticipated.
Anyway, my favorite work for technical details on the Manhattan Project is B. Cameron Reed's The History and Science of the Manhattan Project, which looks at the technical issues closer than Rhodes does.