The Subscriber & Commerce Data Science team at Disney Streaming needs to optimize payment processes, detect and prevent fraud, and forecast customer lifetime value across their streaming platforms to grow the business by increasing payment success, reducing fraud, improving retention, and enabling value measurement through user-level lifetime value (LTV) modeling.
Requirements
- Proficiency in SQL, Python (e.g. Pandas, NumPy, Scikit-learn, LightGBM); experience with distributed computing tools such as Spark or PySpark.
- Deep expertise in statistical modeling and machine learning, including Bayesian methods.
- Familiarity with tools like Databricks, Snowflake, Airflow, GitHub.
- Experience designing and analyzing A/B tests and other experiments.
- Experience with data visualization and exploration tools such as Tableau, Looker
- Ability to choose and justify appropriate modeling and statistical techniques for varied problems.
- 3+ years of experience developing and deploying machine learning models in production.
Responsibilities
- Develop, optimize, and maintain models for payment optimization, fraud detection, and LTV prediction.
- Build robust end-to-end ML workflows, including data collection, feature engineering, model development, and evaluation.
- Collaborate with Product and Engineering to deploy models into production environments and monitor performance.
- Design and analyze A/B tests and other experiments to assess model impact.
- Implement batch and real-time inference pipelines for fraud detection and payment optimization use cases.
- Analyze subscriber behavior, payment flows, and fraud patterns to generate actionable insights.
- Translate complex data into clear, data-driven recommendations to improve business outcomes.
Other
- Bachelor’s degree in Data Science, Computer Science, Statistics, Mathematics, or a related field.
- Partner with stakeholders to translate business needs into machine learning problems.
- Collaborate with Engineering to improve data pipelines, experimentation frameworks, and model monitoring.
- Communicate insights effectively to technical and non-technical stakeholders.
- Comfortable working in fast-paced environments with evolving priorities.
- Excellent communication skills with both technical and non-technical audiences.
