Round 1: Technical [Coding + SQL + ML/DL Theory]
Questions:
- 3 SQL questions (Simple questions queries were using group by, dense_rank and case statement)
- Do the same but if the data is in dataframes instead of tables, get same result using pandas
- DSA Question: Majority Element
- ML questions testing knowledge of concepts, including:
- overfitting
- regularization
- decision trees
- XGBoost
Candidate's Approach
No approach provided.
Interviewer's Feedback
No feedback provided.
Round 2: Technical [ML/DL Theory + Projects]
Questions:
- Deep dive on any one ML project which I handled end-to-end at current organization:
- What variables you used, how many
- If you were told today to build this project again, what would be your approach, broadly tell your steps
- Follow-ups, reasoning, cross questioning on previous answers
- More questions on:
- Boosting
- Bagging
- Decision trees
- Model evaluation metrics
- Cross validation
- Feature importance
- Why leave current organization? Why join PayU?
- Discussion on expected compensation
Candidate's Approach
No approach provided.
Interviewer's Feedback
No feedback provided.
Round 3: Technical [Analytical thinking + ML Case study]
Questions:
- Walk me through your projects in previous organization - explain in short, and what size of data you were dealing with?
- Picked one project and some questions on that
- How do you validate your ML model?
- How do you handle categorical columns?
- Took an example of an internal project and asked few questions to test approach and analytical thinking
- Some behavioral questions
Candidate's Approach
No approach provided.
Interviewer's Feedback
No feedback provided.