The process was long but I learned a lot and everyone I talked to was very kind, helpful, smart, and understanding. I was contacted by a recruiter after applying online and all of my interviews after that were conducted via Google Hangouts. I had an initial screen with someone from HR about what kind of job I was looking for, my general experience, and why I was interested in DeepMind. Next there was a two hour verbal quiz about math (mostly linear algebra and calculus but a couple of other topics too), CS, machine learning, and statistics with four separate interviewers (each specializing in questions about one of the four categories). I did well but didn't get everything right (probably scored around 80-85% as estimated by my panicked post interview calculations...), but I got to the next round despite getting a few questions wrong. The next round was a (virtual) coding interview that started with CS concept questions and then had a typical coding exercise on Google Docs. After this I had several higher level technical discussions with senior research engineers where they asked technical questions about my experience, interests within AI, and papers I had read. Everyone was smart but understanding when I didn't know something exactly, and encouraging. Besides my technical abilities, everyone was very interested in my motivations and why I wanted to work for DeepMind; they really wanted to know that I wanted to work there and was going to do a good job and put my all in if I was hired. At the end of it all I had a final behavioral screen and another high level technical interview and they made me an offer.

Describe your research papers and then a coding questions plus tehcnical and theoretical questions in the field of deep learning relevant to the role. THen questiosns for myself and scenario based and what I would do if i had a problem with one of my coworkers etc