Zepto is a fast‑growing Indian e‑commerce startup that has redefined grocery delivery with its ultra‑quick, app‑first approach. Founded in 2020, the company operates a hyper‑local network of micro‑fulfilment centres across major Indian cities, enabling deliveries in under 10 minutes. Backed by leading venture capital firms, Zepto has scaled to millions of daily orders and is continuously expanding its product portfolio, logistics capabilities, and data‑driven decision‑making framework. The culture at Zepto is built around speed, experimentation, and a relentless focus on customer delight, making it an exciting place for fresh talent to make a tangible impact. The Data Scientist I – Growth role sits within the Growth Analytics team, which partners with product, marketing, and operations to unlock revenue opportunities through data. As a junior data scientist, you will work on real‑world, high‑volume datasets to uncover insights that drive user acquisition, retention, and basket size. You will be mentored by senior scientists and will get hands‑on exposure to the full ML lifecycle – from data ingestion and cleaning, through feature engineering, model building, and rigorous evaluation, to deployment and monitoring in production. Key responsibilities include: 1. Collecting, cleaning, and preprocessing large‑scale transactional and behavioural data from Zepto’s platforms. 2. Designing and implementing feature pipelines that capture user intent, product affinity, and temporal patterns. 3. Building and fine‑tuning tree‑based models (e.g., XGBoost, LightGBM) for churn prediction, recommendation, and demand forecasting. 4. Evaluating model performance using appropriate metrics such as AUC‑ROC, precision‑recall, and calibration curves. 5. Conducting A/B tests and causal analysis to validate the impact of data‑driven interventions. 6. Collaborating with product managers to translate business problems into analytical solutions. 7. Communicating findings through dashboards, visualisations, and concise reports for stakeholders. 8. Monitoring deployed models, diagnosing drift, and iterating on improvements. 9. Staying updated with the latest research in machine learning, statistics, and growth analytics. 10. Contributing to the team’s knowledge base by documenting code, experiments, and best practices. The tech stack revolves around Python (pandas, scikit‑learn, XGBoost), SQL, and PySpark for distributed processing. Version control is managed via Git, and experiments are tracked using tools like MLflow. Zepto encourages a growth mindset – high performers can progress to Senior Data Scientist or specialize in areas such as recommendation systems, pricing optimisation, or experimentation engineering within 2‑3 years. Joining Zepto means working in a high‑velocity environment where data directly influences product decisions, offering a steep learning curve and the chance to shape the future of grocery delivery in India.