Data Scientist – Global Test & Learn Platform

About the role

Contributing to the Data Science efforts of AB InBev’s Global Test & Learn Platform – an analytics engine and SDK that powers experiment design and uplift measurement across multiple markets. The role requires an AI‑first mindset, strong analytical rigor, and the ability to turn business questions into actionable, data‑driven recommendations.

Key responsibilities

• Partner with business, product and analytics stakeholders to define analytical approaches and deliver models that drive measurable outcomes.
• Develop predictive, descriptive and prescriptive models, including hypothesis testing, forecasting, optimization and statistical inference.
• Perform data exploration, feature engineering and model evaluation on structured and unstructured data at scale using Databricks and PySpark.
• Write production‑quality Python code for the platform’s SDK and support its integration with the web UI.
• Build and maintain reproducible, scalable ML pipelines from feature engineering through model serving and monitoring.
• Enhance platform capabilities with LLMs for automated insight generation, AI‑enabled UI features and intelligent summarisation.
• Communicate findings, assumptions and recommendations clearly to technical and non‑technical audiences.
• Support experiment design, A/B testing and causal impact measurement where relevant.

Required profile

• B.Tech/B.E. or Masters in Statistics, Mathematics, Computer Science, Economics/Econometrics or a related quantitative field.
• Minimum 3 years of relevant experience in data science or analytics.
• Experience working with large datasets and cloud or distributed data environments (e.g., Azure, Databricks).
• Experience deploying analytics solutions for internal or external clients, preferably in FMCG, CPG, supply‑chain or commercial analytics.

Required skills

• Python (expert level, end‑to‑end ML pipelines, OOP, packaging, testing).
• Advanced statistics (probability, inference, hypothesis testing, regression, experimental design).
• Machine Learning (regression, classification, clustering, forecasting, optimization).
• Data manipulation and SQL (Pandas, NumPy, SciPy, SQL).
• PySpark and Databricks (intermediate).
• GenAI / LLMs (intermediate – RAG pipelines, prompt engineering).
• Git (intermediate – branching, PRs, code reviews).
• Cloud platforms – Azure (intermediate).
• Familiarity with MLflow, Docker, Streamlit, Power BI, MLOps and experimentation design techniques.