
Each session is a realistic, full-length interview covering the math (linear algebra, probability, statistics, optimization), core ML theory, and the applied depth interviewers actually probe for. You get live pushback during the session and written feedback afterwards: where you were solid, where you were hand-wavy, and exactly what to fix before your real interview.
I've sat through hundreds of these rounds as a candidate across product and consulting companies, and cleared most of them. I know what the bar looks like from the other chair.