Comment by dekhn

This happened when I was in grad school back in the late 90s. Clinton increased the budget of the NIH significantly (thanks Bill) which led to a huge increase in training grants allowing PhD programs to expand. However, faculty positions did not expand, so many PhDs had to find alternate careers (most people go into PhD program to get a faculty/researcher position). This actually worked out really well for bigtech, which hoovered these folks up. I worked with tons of ex-physicists who were great programmers especially at machine learning. It turns out that many of the things you need to thrive in a PhD program translate to bigtech engineering needfs.