Technical interviews asked python and SQL questions. Generally SQL questions involved doing some joins and groupbys. Python questions typically had you read in data and then calculate some quantities. Sometimes they let me use Pandas and other times they had me basically re-create a pandas function which seemed a little strange. Was asked a few specific questions about time-series modeling due to the position.
Describe how regularization works for linear models
Describe a time when you encountered an obstacle. How did you deal with it?
What is the assumption of error in linear regression?
Thorough hands on data science workflow (eda, data cleansing, ML application/ feature engineering iteration) to solve a business problem.
DSA question with mL GenAI
Case study: new product scenario and how will you define the metrics to track. SQL: self join, time functions, and window functions.
What metrics would you use to measure the success of a new product launch? How would you calculate churn rate, and what key metrics would you monitor? If tasked with improving user engagement, which metrics would you focus on and why?
Here's an ugly data set. Do exploratory analysis. Create visualizations. Discuss. Build a predictive model with different approaches, concerns,validity, performance...
Basic business case, how would you approach the problem, what data would you use, what model, how would you validate it
Viewing 1891 - 1900 interview questions