There was an initial screening call. After, there was a 30-minute technical call. The last step was an onsite 4-hour long interview. The first part of the onsite interview was speaking with the directors of research for 45 minutes, which was then followed by a 30 to 45-minute presentation. The presentation can be a candidate's thesis, class project, etc., as long as it's work within data science—my PhD thesis is in theoretical math, so they were perfectly happy with me presenting on a class project instead—be prepared to answer some pretty probing questions. The penultimate part is to go out for lunch with a current employee, followed by a 30-minute meeting with the VP. As a candidate, I felt I was given a fair chance to show who I am and what I can do. Everyone I interacted with, via email or in person, was kind, professional, and very helpful. This was by far my best interview process. I had my answer ready the day they gave me an offer.

Phone interview with HR, then a primary phone interview with a program manage, and lastly a full day interviews with seminar talk, analytical review, and one-on-one interviews in the afternoon.