 Statistical inference by Kasella and Berger. It is hard to understate the value of this book. Even if all someone reads is a first-do chapters that would be sufficient for any entry-level position in data science. The problem set is fun and challenging and you will probably never get through all of it. In my master's program I'm not sure if we even did half the problems and that was with two semesters dedicated to this book. There are a number of interesting tables like the table in the back of the book which shows how many of the popular probability distributions are related. This is a graduate-level textbook. It is a must-have for anyone trying to learn mathematical statistics. The book is heavy on definitions and expects the reader to play with those definitions in the form of applications, derivations, and occasional proofs. Though I've been out of school for many years I will still pick this book up, pick the section at random, and read it through. It keeps me sharp and even if I read a section before I often learn something new on a third or fourth pass. There is a reason this continues to be a recommended book for those looking for rigorous mathematical statistics.