Data Scientist

The role focuses on translating messy, real-world data into models and analyses that drive core product and business decisions - not dashboards that sit in a slide deck, but statistical and ML work that ships to production.

The role focuses on translating messy, real-world data into models and analyses that drive core product and business decisions - not dashboards that sit in a slide deck, but statistical and ML work that ships to production. You will partner closely with ML engineers and product owners to frame problems, run rigorous experiments, and build models spanning forecasting, causal inference, and customer behavior prediction, with full ownership from hypothesis to deployment.

• Design and execute statistical analyses and experiments (A/B tests, quasi-experimental designs) that inform product roadmap and pricing decisions
• Build, validate, and ship predictive models - forecasting, propensity, churn, ranking - using Python, scikit-learn, and gradient boosting frameworks (XGBoost, LightGBM)
• Develop production-ready data pipelines in SQL and PySpark, partnering with data engineering on feature stores and data quality standards
• Frame and solve causal inference problems using difference-in-differences, propensity matching, or uplift modeling where A/B tests aren't feasible
• Communicate findings and model tradeoffs directly to engineering and product leadership, including uncertainty, limitations, and business impact
• Transition proven models to ML engineering for deployment, defining monitoring metrics and retraining criteria
• Contribute to team standards around experimentation methodology, model documentation, and reproducibility

• 3–6 years of experience in data science, applied statistics, or ML, with multiple models taken to production
• Advanced Python skills (pandas, Num Py, scikit-learn) plus strong SQL against large-scale data warehouses (Snowflake, Big Query, or Redshift)
• Hands-on experience designing and analyzing controlled experiments, including power analysis, variance reduction, and guardrail metrics
• Solid grounding in statistical fundamentals: hypothesis testing, regression, regularization, and common failure modes of observational data
• Experience with at least one distributed data framework (PySpark, Databricks) and cloud environments (AWS, GCP, or Azure)
• MS or PhD in a quantitative field (Statistics, CS, Economics, Math) or equivalent practical experience
• Bonus: causal inference tooling (DoWhy, EconML), LLM-assisted analytics workflows, or experience building internal experimentation platforms