Masters Candidate, Tier1 College
Campus Hiring

Role: Data Scientist II Intern
Duration of Internship: 6 months
Stipend: 1.4 LPM
Number of Rounds: 2
Round 1: Standard Coding Test (7 Debugging Questions + 2 Coding Questions + Behavioral Test + Logical Reasoning)

Solved 7/7 debugging questions. The coding questions were easy - 1 solved with 22/22 test cases passed, other with 13/16 test cases passed. I must have done well (relatively) in the coding test which is why I was the first candidate to get interviewed.

Round 2: Interview (Online) on Amazon Chime
Interview Questions:

1. Assumptions of Linear & Logistic Regression?
2. What variations of Gradient Descent (Batch GD/Stochastic GD/Mini-Batch GD) will you use for each of the following 4 scenarios?
   2.1. Less number of examples, less number of features
   2.2. Less number of examples, more number of features
   2.3. More number of examples, less number of features
   2.4. More number of examples, more number of features
3. How does the regularizing parameter in Linear Regression help in preventing underfitting and overfitting?
4. Given two binary classification models, one with AUC of 0.6 and one with AUC of 0.3, which one is better? (Answer – 0.3)
5. What is the physical significance of Precision and Recall?
6. Elaborate on the loss function of a deep learning project that you have worked on?
7. Can you tell the mathematical expression for SVM loss?
8. Comment on the bias and variance for underfitting and overfitting scenarios?
9. Explain bagging? Comment on the bias and variance for a Decision Tree and Random Forest fitted on the same data?
10. Explain the problem of vanishing/exploding gradients?
11. What are the two steps in K-Means clustering?
12. How to find the Kth largest number in an array?

I was selected after this round. I had answered pretty much all questions (except 1-2 partially). If you want me to answer any of the above questions, please let me know and I will be happy to answer it.

Additional Note: The coding round carries a significant weightage because I was considered as High Importance candidate. I can say this because the Placement Representative forwarded (perhaps by mistake) an email with the Chime link from the HR and the mail contained subject as Importance: High.

Hope this helps.