First Round: On campus Interview. 45 mins. Two probability questions, one programming question. General questions about machine learning algorithms. One on one interview. Relaxed and comfortable, guide you through the process. White boarding. Programming language: choose the one you feel comfortable with. I chose Python. Basic probability questions concerning combinatorics and conditional probability.

Recruiter contacted me via email I think through college and university. Initial phone interview consisted of a couple regular behavioral questions and some technical. Asked what's a confidence interval and what does it represent. Asked to open up a pair programming website and asked how to merge two sorted arrays. Next step was an on-site interview. This was a whole days worth of thirty minutes each with a different interviewer. Each interviewer jumped into the technical questions immediately with absolutely no room for chit chat. Very different from other Workday interviews on Glassdoor that are more culture fit because of the use of learning Expresso, you get the sense that the data science team is exempt from that and is testing really for pure skill. Questions: Statistics and Machine Learning - AB testing on a website - Anova testing - Explain naive bayes, map reduce, cross validation, how would you tag specific documents - Boosting question given an array Coding: - Find the top m words in an array - Find multiplication pairs for a specific n value - Sorting an array of integers Overall a good experience although it took a long time for them to get back to me.