Overview
Apply statistical methods and machine learning across the full model lifecycle to improve customer retention, growth, and commercial effectiveness, turning customer, product, and commercial data into machine learning solutions that inform business actions.
What you'll do
- Translate business problems into modeling approaches for use cases such as churn, upsell, cross-sell, lead and prospect propensity, product recommendations, and price elasticity.
- Conduct exploratory data analysis to investigate distributions, missing data, customer segments, and feature relationships; assess data quality and prevent target leakage.
- Design reproducible experiments, compare baselines and candidate models, and validate results using appropriate sampling, cross-validation, temporal testing, and business-relevant performance metrics.
- Diagnose model issues by investigating prediction errors, feature behavior, calibration, stability, and Shapley additive explanations (SHAP) to identify improvements.
- Work with engineering teams to deploy, test, monitor, and retrain models; document assumptions, communicate uncertainty, and measure business outcomes.
What you'll need
- A master's degree in computer science, data science, artificial intelligence, statistics, or a related quantitative discipline, with academic grounding in mathematics, statistics, computer science, or another quantitative science.
- Applied knowledge of probability, statistical inference, hypothesis testing, regression, classification, experimental design, and model evaluation, including bias–variance tradeoffs and uncertainty.
- Experience using Python, structured query language (SQL), and machine learning libraries to prepare data, engineer features, and build reproducible modeling workflows.
- Experience developing and deploying predictive models in business-to-business or business-to-consumer companies, including validation, production monitoring, and troubleshooting, with examples of business application and measured results.
Nice to have
- Experience with several commercial use cases, such as retention, purchase propensity, recommendations, or pricing.
- Peer-reviewed publications, a research thesis, or equivalent applied research demonstrating methodological rigor and the ability to evaluate scientific evidence.
- Experience with causal inference, uplift modeling, imbalanced datasets, or probability calibration.
- Experience with version control, automated testing, experiment tracking, and model deployment practices.
Details
- Location: Bangalore, India.
Read the full description and apply on the company’s own careers page.