Phone screen 1. If given a univariate data, how would you design a function to detect anomalies? 2. If the dataset now is bivariate, how would you proceed with creating the function to detect anomaly? 3. How would you compare two categorical variables? 4. What would the results of a chi-squared test look like? On-site (there were 5 rounds with 5 different people- data scientists, hiring manager, engineering manager and director, the questions are summed as below) 1. Given a log file, how would you detect the most number of agents (the file contains- IP Adress, agent, timestamp and a couple of more values) 2. What is the probability of two students having birthday on the same day? 3. Describe what is clustering algorithm and give an example 4. How would you solve a class imbalance problem? 5. What is a sigmoid function, draw and explain 6. What is PCA, explain? 7. What is overfitting? 8. Given an array of integers and a target number, find the two indices in the array that will sum up to the target number. Code on whiteboard 9. The same question as above was modified to finding three indices in the same array. Code on whiteboard 10. What was one project that you failed and why do you think so that it failed? How do you think you would have improved it? 11. Explain a recent project in detail (more business oriented and focussed on thought process) 12. Did you face challenges in working with a teammate in the past and why? 13. Why do you want to quit your current job? 14. Live coding in python to load a txt file, find maximum, minimum points, finding unique values in the file Tips- Use whiteboard maximum number of times to explain your approach