> What do your grad students do about the fact that for whatever reason, they are largely not headed for the tenure track? How do you advise them about their ca

> What do your grad students do about the fact that for whatever reason, they are largely not headed for the tenure track? How do you advise them about their careers? It depends a lot on the student. I've had 3 Ph.D. students graduate already (and 1 grad student/post doc), and they all now have good jobs in industry (e.g., 2 as software engineers at Google). I also have several students right now. My students often do a lot of work on SageMath (http://sagemath.org http://sagemath.org), so at least get relevant experience that makes them employable outside pure math academia. I also encourage them to get involved with or start other open source projects (example: http://cython.org http://cython.org), which might have basically nothing to do with their thesis, but which makes them more employable. In short: 100% of my students leave grad school with tons of experience doing open source software development, in addition to whatever else they do. I do have a student right now who has turned out to be very good at number theory research, and wants an academic job -- it will be interesting to see if he gets one... > My advisor accepted my decision, but admitted that he had no advice for me. [...] One thing he did, to his credit, was to start bringing in all of his former students who were in industry, to give talks. I go out of my way to try to understand something about the world outside academia, so at least I don't have 0 advice for my students. It's great that your adviser did something that he could to help. > Another question: If your students are headed into job markets where publications don't matter, how do you maintain your own publication record? Surprisingly, most conventions involving publication in pure mathematics are somewhat different than in much of the rest of the academic sciences. For example, pure mathematicians always list authors in alphabetical order. Pure mathematics is also unlike most of the sciences, because professors write their own papers. There are exceptions of course, but usually you can only be a co-author on a paper if you actually write part of it (!). Grad students do not write papers for their advisor (modulo rare exceptions). Last week I had one of my grad students get a paper accepted in a very good journal on things I met with him every week about, and I'm not a co-author. With this paper (http://wstein.org/papers/bs-heegner/ http://wstein.org/papers/bs-heegner/) the student went to work at Google right after graduating and stopped doing math, so I had to do the work of writing up for publication and submitting the paper, just to ensure it got published. Frankly that's just fine; if that paper would have been 100% written up and submitted by the student, I definitely wouldn't have been a co-author.