My main point in response to the post from the Google recruiter is that Google's recruiting process is messed up, say, arrogant, inwardly directed, process orie

My main point in response to the post from the Google recruiter is that Google's recruiting process is messed up, say, arrogant, inwardly directed, process oriented, too narrow and particular, bending over backwards to find silly reasons to reject people, and with irony, actually not "prepared". My evidence is (1) the emphasis in the post on Skienna and the contents of that book, (2) the claim about lack of knowledge of dynamic programming meant not "prepared", (3) the common complaints about the Google process emphasizing tricky questions, and (4) my experience where, with irony, if they like the topics in Skienna, I certainly should have done well in the interview but didn't. "You know that after reading only the table of contents?" Sure: I know nearly all the topics quite well. For the depth of coverage of the book of each of the many topics, can conclude that the depth is shallow because of the wide variety of topics in the book, a 'catch all', the few number of pages for each topic, and the lack of more table of contents outline details for the topics. E.g., in linear programming also need to discuss slack and surplus variables, artificial variables, feasible, infeasible, unbounded, bounded, optimal, basic feasible solutions, the simplex algorithm with reduced costs, a pivot rule, entering variables, leaving variables, convexity, extreme points, degeneracy, cycling, and performance. So, for much on linear programming, will need more than one entry in the table of contents. "You worked much more deeply the topics in a book that you didn't read or even see? And that's an argument for what? What relevance this has to anything?" It says (1) I'm qualified to claim that that book is a poor means of preparation for a good interview in an appropriate recruiting effort and (2), with irony, if Google actually likes the topics in that book, since I know the topics well, I should have done well on the interview but did not due to some Google recruiter's religious worship of C++. "Good for you, I guess. Again, what relevance this has to the discussion?" Again, the "relevance" of my knowledge of the topics in the book means that I have some qualifications to comment on the relevance of that book for interview preparation. My points here are that (1) Google should not be recommending a particular book, since really there are many sources including many much deeper than that book, but, perhaps, recommending a list of topics instead and (2) they should not be asking about dynamic programming. "Because you did research you now want everyone to spend at least a year learning optimization?" Heck no: I'm saying that mostly software developers should not be studying optimization and that a Google interview should not be asking questions about optimization at the level of that book. Mostly they shouldn't be asking about optimization at all. If, in some particular case there is a good reason for knowledge of optimization, then ask about the subject in a serious way. Again, the claim that someone needed to know dynamic programming at the level of that book or was not "prepared" is absurd. At the level of that book, f'get about dynamic programming. "You want all books to cover only advanced optimization and not the basic stuff?" You won't find anything like good coverage of the "basic stuff" about optimization in that book. It's a very old story: There was a chemistry book that wanted to introduce group representation theory because it plays a role in molecular spectroscopy important for identifying chemical molecules. So the book had a few pages on group representation theory. They had a mess. The chemists came to the math department for some help. I ended up writing my honors paper on group representation theory. The lesson is, people in field B should not write short introductions to topics in field A. Instead, if want a short introduction to field A, then get it from someone actually in field A. It's important. Otherwise too often end up reading junk as in that chemistry book. That book is awash in topics in applied math outside of computer science. The chances that the content is good, even for its short length, are slim to none. There is a pattern in computer science: It keeps trying to borrow from other fields and, then, makes a mess out of the content of the other fields. A big example is how computer science borrowed 'machine learning' from statistics and made a mess. More generally, nearly none of the profs in computer science are qualified to write about math at all. Sorry 'bout that. "You want Google to hire only people who took a graduate-level optimization course?" No. Except for a position that is clearly in optimization, Google should just f'get about optimization. "It's hard to read this combination of arrogance, self-promotion and name-dropping, which in the end hardly adds up anything to the discussion." What I "add" to the discussion is that Google is making a mess of their interviews. The "name dropping" is to establish my qualifications for saying that Google should mostly just f'get about optimization. The "arrogance" is Google saying that lack of knowledge of dynamic programming means not "prepared". Then there is the irony: Apparently Google is not "prepared" in dynamic programming in any meaningful sense. For your last paragraph, the original post did make Skienna look like the "official Google knowledge base" for job interviews. "From what I understand they simply ask you some basic algorithm questions at Google and I guess that's pretty natural." No: The original post implied that a candidate should have reviewed the first half of Skienna including dynamic programming, and my view, from good knowledge of that material, is that, then, Google is doing poor interviewing.