Senior Data Scientist – Experimentation & Causal Inference
Company: Umanist NA
Location: Glendale, CA (Remote)
Salary: $135k - $140k per year
Type: Full-time
Remote: Yes
Posted: 2026-07-02
About this role
Technical Responsibilities
- Design and Execute Experiments: Lead end-to-end A/B testing initiatives and Geo Experiments, from hypothesis formation and experimental design to statistical analysis and business recommendations.
- Advanced Statistical & Causal Inference: Apply deep knowledge of experimental design, regression, classification, causal inference (difference-in-differences, propensity scores, instrumental variables), and ensure proper assumptions.
- Build Scalable Solutions: Develop experimentation and causal inference tools and frameworks that can scale across Disney's businesses.
- Deliver Strategic Insights: Partner with stakeholders to identify optimization opportunities and translate complex analytical findings into clear business recommendations.
- Influence Executive Decisions: Present findings and recommendations to senior leadership, effectively communicating statistical concepts to non-technical stakeholders.
Basic Qualifications
- Bachelors in Statistics, Economics, Computer Science, Engineering, Mathematics, Physics, or a related field + 7 years of experience with an emphasis on experimentation or causal inference.
- Strong background in statistical modeling: regression, classification, time series forecasting, causal inference, and other techniques.
- Robust knowledge of causal inference approaches such as propensity scores, synthetic controls, difference-in-differences, doubly robust methods, meta learners, and uplift modeling.
- Expertise in A/B test design, execution, statistical modeling, and sophisticated causal inference techniques.
- Proficient in conducting sample size calculations, power analysis, and minimum detectable effect estimation.
- Experience managing multiple testing scenarios and controlling false discovery rates.
- Ability to deploy both Bayesian and frequentist statistical approaches.
- Deep understanding of assumptions required for causal inferences, including the foundational statistical concepts that underpi...