Swiggy

Role: Data Scientist 1

College: Not Specified

Interview Date: 2024

Interview Type: Not Specified (Leveraged a referral)

Result: Offer

Difficulty: Medium to Hard

Rounds: 2 rounds (after CV shortlisting)

Topics Asked: Current Work, Business Impact, Markov Chain Probability, Gradient Boosting Trees (GBT), AUC-ROC Curve, L2 Norm, Cosine Similarity, Recommender Systems, Case Study, Models, Evaluation Metrics

Experience:

The candidate secured an interview through a referral, with their CV being shortlisted based on prior work in Data Science, particularly GBT and Recommender Systems.

Round 1 (Interview with Data Science Team Lead): This round focused on the candidate's current projects and their business impact. Technical questions included Markov Chain Probability concepts, detailed questions about GBT (residual calculations, hyperparameters), AUC-ROC Curve (understanding and evaluating models), L2 Norm, and Cosine Similarity.

Round 2 (CV Review and Case Study): This round involved an in-depth review of the candidate's CV, covering models and evaluation metrics mentioned. A case study was presented, requiring the candidate to solve a recommendation system problem focused on ad matching and optimizing user engagement.