Senior Data Scientist – Marketing Mix Modeling (MMM)

As a Senior Data Scientist – Marketing Mix Modeling (MMM), you will play a critical role in designing and delivering advanced Bayesian Marketing Mix Modeling solutions that help clients optimize marketing investments, improve media effectiveness, and drive measurable business outcomes. You will develop statistical models, causal inference frameworks, and optimization solutions that enable data-driven marketing decisions across global enterprises.

This role requires deep expertise in Bayesian statistics, probabilistic programming, causal inference, optimization, and marketing analytics. You will collaborate closely with cross-functional teams, including data engineers, product owners, business stakeholders, and client leadership, to build scalable, production-ready analytics solutions.

Key Responsibilities

• Design, develop, and deploy advanced Marketing Mix Models (MMM) using Bayesian statistical methodologies.
• Build Bayesian regression models with appropriate likelihoods, priors, and hierarchical model structures.
• Develop probabilistic models using PyMC and/or Stan for marketing effectiveness measurement.
• Design and implement hierarchical Bayesian models to improve estimation across sparse or multi-level datasets.
• Develop chained and multi-stage modeling frameworks with proper uncertainty propagation using Monte Carlo simulations.
• Build optimization frameworks for marketing budget allocation using constrained nonlinear optimization techniques.
• Develop multi-objective optimization solutions balancing ROI, business constraints, and marketing objectives.
• Design and evaluate causal inference frameworks using geo experiments, Difference-in-Differences, Synthetic Control, and other quasi-experimental techniques.
• Implement adstock transformations, saturation functions (Geometric, Weibull, Hill curves), and response curve estimation for media effectiveness.
• Perform model diagnostics, posterior analysis, convergence validation, and uncertainty quantification using ArviZ and Bayesian diagnostic tools.
• Collaborate with business stakeholders to translate analytical findings into actionable marketing recommendations.
• Develop production-quality Python code, reusable modeling frameworks, and reproducible analytical workflows.
• Document modeling methodologies, assumptions, validation processes, and technical findings.
• Mentor junior data scientists and contribute to technical best practices across the organization.