While preparing for a data science job interview at Microsoft, candidates need to prepare about topics ranging from analytical thinking to problem-solving. In this article, we list down some of the top data science questions to ace your interview as well as some important tips to prepare for D-day.\r\n\r\nIn a blog post by Microsoft, the recruiters shared that they expect the candidates to have a passion for coding, solving problems and the incredible things technology can do for people around the world. Every interview looks for something specific depending upon a particular job profile.\u00a0\r\nTechnical Interview\r\nA technical interview varies depending upon the team and the role which the candidate is going to apply for. Some of the important points to be noted while appearing in a data science interview are:\r\n\r\n \tA Neat Resume: In an interview, a resume acts as a mirror for your experiences and achievements. A resume must be neat enough to understand. Also, the candidate must include real-life examples, projects, and experiences in the resume by depicting its impact as well as how it benefits the end-users. Before an interview, one must also go through all the topics which are mentioned in a resume.\r\n \tCommunication Skills: Interviewers are no mentalists. This is a general point where strong communication is needed while appearing for an interview. Also, while responding to any questions, the candidate must understand what is being asked by the interviewers and respond accordingly.\r\n \tDelving Deep Into The Basics: In a data science interview, a candidate must dive deep into topics such as the basics of machine learning algorithms, statistics, linear algebra, probability, among others. There is a plethora of online courses where one can easily brush up the basic concepts beforehand.\r\n \tThe Code To Code: While appearing in a personal interview round, the interviewer might ask you to solve a coding problem. The right way to approach it is to ask any doubt or questions related to the problem and not just directly rush into solving the problem. Following this rule will not only help you to complete the task with a more accurate answer but also it will showcase your communication skills and the way of understanding and solving a problem.\u00a0\u00a0\u00a0\r\n\r\nSolve The Questions\r\nNow, we will take a look at the questions which are usually asked in a data science interview at the tech giant and preparing them will surely help you to ace your data science interview.\r\n\r\n \tGiven a string like \u201cfun1(fun2(a,b) ,c ,fun2(a),fun3(c,d))\u201d how many unique function signatures are there.\r\n \tYou have a bag with 6 marbles. One marble is white.\u00a0 You reach the bag 100 times. After taking out a marble, it is placed back in the bag. What is the probability of drawing a white marble at least once?\r\n \tGiven a box of dimensions W, H, and coordinates of points inside that box. Find the largest area that is free of any of these points.\r\n \tWhy neural network work and why is it a booming field?\r\n \tWhat's power? How to explain it to a non-statistics person? what's a false positive and false negative?\r\n \tHow Random Forest, Lasso and Ridge Regression work? Difference between lasso and ridge.\r\n \tWrite a function to check whether a particular word is a palindrome or not.\r\n \tFind the maximum of subsequence in an integer list.\u00a0\u00a0\r\n \tGenerate a fair coin from a biased one.\r\n \tGenerate 7 integers with equal probability from a function which returns 1\/0 with probability p and (1-p).\r\n \tWhat are the ROC curve and the meaning of sensitivity, specificity, confusion matrix?\r\n \tGiven a time series dataset, how will you predict future value?\r\n \tHow to explain a deep learning model to customers?\u00a0\r\n \tWhat is the definition of a P-value? How to explain p-value to customers.\r\n \tHow can you compute an inverse matrix faster by playing with some computation tricks?\r\n \tDescribe how gradient boost works.\r\n \tDescribe the steps for data wrangling and cleaning before applying machine learning algorithms.\r\n \tHow to deal with unbalanced binary classification?\r\n \tHow do you detect if a new observation is an outlier? What is the bias-variance trade-off?\r\n \tExplain the Support Vector Machine (SVM).\r\n\r\nWrapping Up\r\nThe interview questions above have been cited from various sources, comments and reviews about data science interviews at Microsoft. Also, we can see that the questions are mostly related to probability, basics of machine learning algorithms, and other techniques of extracting data in an ML method. Thus, a candidate with a good grip on these topics will surely ace the data science interview.