Data Scientist

Role Overview

We are seeking a highly skilled expert to develop, implement, and optimize Marketing Mix Models (MMM) to measure the impact of marketing investments and support strategic budget allocation. This role involves building robust Bayesian statistical models, applying causal inference methodologies, and designing advanced statistical modelling techniques to drive marketing effectiveness and incrementality measurement.

Responsibilities

• Develop and optimize Marketing Mix Models (MMM) for budget allocation decisions.
• Build Bayesian statistical models for forecasting and uncertainty estimation.
• Apply causal inference to distinguish correlation from causation in marketing campaigns.
• Design advanced techniques including regression, hierarchical Bayesian models, and time-series analysis.
• Execute hypothesis-driven experimentation such as A/B testing and geo experiments.
• Build scalable Python-based analytics pipelines for model development and monitoring.
• Collaborate with cross-functional teams to translate findings into optimization strategies.
• Build automated dashboards using Power BI or Looker Studio.

Requirements

• 3–6 years of experience in Marketing Analytics, Marketing Science, Applied Data Science, or Econometrics.
• Strong experience in agency, consulting, or digital marketing analytics environments.
• Expert knowledge of Marketing Mix Modelling (MMM) and Bayesian Inference.
• Deep expertise in Statistical Modelling (Linear, Multivariate, Hierarchical, Time-Series).
• Hands-on experience with Causal Inference (Difference-in-Differences, Synthetic Control, etc.).
• Proficiency in Python (pandas, NumPy, SciPy, scikit-learn, PyMC, Statsmodels) and SQL.

Nice to Have

• Experience with Google Meridian or frameworks like LightweightMMM and Robyn.
• Experience with GCP, BigQuery, or Vertex AI.
• Knowledge of MLflow, Airflow, Docker, and CI/CD.
• Familiarity with Generative AI for reporting automation.

Skills

• Python
• SQL
• Bayesian Inference
• Marketing Mix Modeling
• Causal Inference