Research

My research focuses on understanding and influencing customer behavior in digital marketplaces, with a particular interest in decision-making, engagement, and long-term value.

I recently completed a PhD in Operations Management at McGill University. My dissertation involved designing and running randomized field experiments with a subscription-based meal kit service to study how different types of promotions; in timing, framing, customization and structure, affect customer retention and behavior over time.

I used techniques from causal inference, optimization, and reinforcement learning to build models that not only explain customer responses but also recommend when and how to act.

What I’ve worked on

  • Promotion design and timing optimization
    How and when to offer incentives for sustainable customer engagement

  • Reinforcement learning for marketing strategy
    Using Q-learning and multi-agent setups to balance short- and long-term goals

  • Causal inference in field experiments
    Uplift modeling, heterogeneous treatment effects, and difference-in-differences

  • Customer lifetime value (CLV) prediction
    Forecasting retention, churn, and order value using behavioral and contextual data

  • Menu recommender systems for retention
    Built hybrid recommenders combining behavioral signals, product metadata, and image features

Methods and tools I use

  • Randomized Controlled Trials (RCTs)
  • A/B testing and uplift modeling
  • Mixed-Integer Linear Programming (MILP)
  • Q-learning and explainable RL
  • Python, SQL, Spark, PyTorch, Scikit-learn
  • Gurobi, Tableau, GitHub Actions, AWS

Publications & Presentations

This research sits at the intersection of theory and practice, shaped by real-world complexity, but grounded in clarity and care. If you’d like to connect or collaborate, feel free to get in touch.