So, while large part of what I am going to write is generic and hopefully useful for everyone, some parts may be specific to that background.Generally, an in-person interview for a data scientist role involves meeting a couple of people:fellow data scientists, of different experience levelssoftware engineers, data engineers etcdata science manager(s), and any other personnel (e.g., in a startup, this can involve a CTO or CEO).Depending on the company, this usually involves about half a day of interviewing..I did talks only when I applied for research focused roles — so, I am not going into that in this post.The moment I knew I am going to apply for data scientist positions, I started a live document, adding questions that I found online, and making my own classification..While this classification may not work for all, here is what I used:questions related to resume/past workprogramming and software engineering questionsmachine learning/Deep learning/NLP: conceptual questionsproblem solving questions (mostly related to company’s domain)statistics questionsbehavioral questionswork culture related questionsquestions we can/should ask the interviewerI combed through various websites, and started listing all the questions I found in my live document, under the appropriate section..Best way to address these in my view is:practice a “tell me about yourself” pitch lasting 1–1.5 minutesthoroughly go through your resumequickly browse through your public profiles (LinkedIn, Medium, Github etc)Questions related to programming and software engineering practices:These are primarily questions you could expect from engineers, and to some extent, from your future colleague data scientists..But if you are totally rusty, it is useful to spend more time, if you can.for software engineering practices, since I am not being interviewed for a traditional software engineer position, I did not spend a lot of time, other than listing down a few questions I would have asked if I was the interviewer (e.g., when do I use notebooks? do I use classes in Python? how do I version data and models? how do I view software testing for data science? etc.)ML/DL/NLP: Conceptual questionsThese are the questions that the data scientists in the team may ask you..In my interviews, these questions typically came from a manager, although sometimes, a senior data scientist also asked these.An example could be something like: “We are a e-commerce website, and we have this problem of fake reviews and fake user profiles..There are a lot of online resources on these, and I combed through as many as possible and listed the unique questions in my document, with answers where I knew them or found them online.Behavioral questionsBehavioral questions are primarily focused on understanding how we react to different scenarios at work place, and know your general attitude..With managers and others, I also want to know if I am sent to any conferences, whether I can get a few days off to do a personal project relevant to the company’s interests etc.Questions I asked as an interviewerWhile I did not spend a lot of time, I did take part in interviewing for data scientist positions..I did not ask all of these, but considering that I am a data scientist, and I had a little bit of engineering and research background in the past, my questions were primarily of four kinds:technical questions: e.g., about some NLP/ML/DL algorithms and their use cases..I did not ask any programming questions as it was usually covered in a online test or by another interviewer.problem solving: e.g., “how will you design a automatic tagging system for medium articles?”questions on resume/past work, processes and practices: e.g., “do you practice Team Data Science Process?”, “Do you do code reviews for other data scientists?”, “in company X, what libraries did you use?” etc.work culture questions: e.g., “what newsletters/blogs do you follow to stay in touch in the latest developments?”It is a rather long post, and all of these were not done in a day or two.. More details