Data Scientist (Pepsi Project)

Role Summary

Data Scientist with 5-10 years of experience having strong Python coding skills, Machine Learning algorithm expertise, Time Series Forecasting experience, and the ability to independently drive model development, deployment, and implementation with minimal supervision.

Key Skills Expected

- Python (NumPy, Pandas, Scikit-learn)
- Machine Learning Algorithms
- Time Series Forecasting
- Model Development and Validation
- Feature Engineering
- Statistical Analysis
- Data Visualization
- Model Deployment & Monitoring
- Stakeholder Communication
- End-to-End Project Ownership

Medium-Level Interview Questions

1. Explain the difference between supervised and unsupervised learning.
2. How do you handle missing values in a dataset?
3. What is overfitting and how can it be prevented?
4. Explain bias-variance tradeoff.
5. What evaluation metrics would you use for regression problems?
6. How does Random Forest work?
7. What is feature engineering? Give examples.
8. How do you select important features for a model?
9. What are the assumptions of linear regression?
10. Explain cross-validation and its perks.
11. What is stationarity in time series forecasting?
12. What is the difference between ARIMA and SARIMA?
13. How would you forecast monthly sales data for Pepsi products?
14. Explain the difference between MAE, MSE and RMSE.
15.



How would you identify and handle outliers?

Expert-Level Interview Questions

1. Design an end-to-end demand forecasting solution for a global beverage company.
2. How would you handle intermittent demand forecasting?
3. Explain Prophet, ARIMA, XGBoost, and LSTM for time-series forecasting and when to use each.
4. How would you detect data drift and model drift in production?
5. Describe a machine learning model you deployed at scale and challenges faced.
6. How would you forecast demand across multiple geographies and product hierarchies?
7. What techniques would you use for feature selection in high-dimensional datasets?
8. How do you optimize hyperparameters for large-scale ML models?
9. How would you build a recommendation or promotion effectiveness model for Pepsi?
10. Explain SHAP values and model interpretability techniques.
11. How would you handle concept drift in a forecasting model?
12. What statistical tests do you use to compare model performance?
13. How would you architect a real-time ML pipeline?
14. Explain ensemble learning strategies and their advantages.
15. How do you ensure reproducibility and governance in ML projects?

If you are interested in this opportunity, please click on the Apply Now button to submit your application.

📌 Data Scientist (Gujarat)
🏢 Tecnoprism
📍 Gujarat