About The Role
The role sits at the intersection of applied statistics, machine learning, and product strategy — building models and analytical systems that directly shape how the business acquires, retains, and serves customers.
This is not a dashboard-only role. The team ships data science to production: experimentation infrastructure, forecasting models, and causal inference pipelines that executives and engineers rely on daily.
Key Responsibilities
Design and run large-scale A/B tests and quasi-experimental studies (difference-in-differences, synthetic control, CUPED) to measure causal impact of product changes
Build and deploy predictive models — churn, propensity, forecasting, ranking — using Python, scikit-learn, and gradient boosting frameworks like XGBoost or LightGBM
Own the full modeling lifecycle: problem framing, data exploration, feature engineering, model selection, and post-deployment monitoring
Develop and maintain SQL pipelines and dbt models working closely with data engineering to ensure clean, reliable training data
Translate ambiguous business questions into rigorous analytical frameworks and communicate findings to senior leadership with clear, quantified recommendations
Contribute to the experimentation platform and statistical methodology standards used across the organization
Prototype and validate new modeling approaches in notebooks, then productionize them with engineering partners
What We Are Looking For
3–6 years of experience in data science, applied statistics, or ML research, with at least one model taken from prototype to production
Expert-level SQL and strong Python (pandas, NumPy, scikit-learn); comfort working with large datasets in Spark or BigQuery
Deep grounding in statistics and causal inference: hypothesis testing, power analysis, confidence intervals, observational study design
Hands-on experience designing and analyzing controlled experiments at scale
Strong data storytelling skills — able to present technical findings to non-technical stakeholders and influence decisions
MS or PhD in a quantitative field (statistics, economics, computer science, operations research) or equivalent practical experience
Bonus: experience with uplift modeling, Bayesian methods (PyMC, Stan), time-series forecasting (Prophet, ARIMA), or building internal experimentation tooling