Sr. Research Data Scientist

Roku•Boston, MA

1d•$330,000 - $375,000•Hybrid

About The Position

Roku is the #1 TV streaming platform in the U.S., Canada, and Mexico, and we've set our sights on powering every television in the world. Roku pioneered streaming to the TV. Our mission is to be the TV streaming platform that connects the entire TV ecosystem. We connect consumers to the content they love, enable content publishers to build and monetize large audiences, and provide advertisers unique capabilities to engage consumers. Our Data Science team is a high-impact research team actively shaping the future of TV, using Big Data to build and enhance the user experience on the Roku streaming platform. Our production-ready machine learning models and statistical solutions optimize the user experience across all of Roku's core business models and products, and our scientists engage closely with business, product, and engineering leaders to make material and measurable impacts on the success and growth of the platform. As a Senior Research Data Scientist on Roku's Data Science team, you will lead the development of a best-in-class causal inference platform that measures and optimizes the true incremental impact of customer actions, product features, and business interventions on long-term outcomes. Partnering with the Customer Growth organization, you will build the methods and systems that enable Roku to make high-confidence decisions from observational data when randomized experiments are not feasible. You will own the full lifecycle of causal measurement—from gathering business requirements and defining estimation approaches, to partnering with Engineering to productionize scalable causal pipelines and communicating findings to senior leadership. Your work will directly inform growth, retention, and monetization strategy across the platform, making this role ideal for an applied economist or econometrician who excels at the intersection of rigorous research and production engineering. This is someone equally comfortable deriving identification strategies and building estimators on terabyte-scale data.

Requirements

PhD in Economics, Econometrics, Statistics, or a closely related quantitative field with a strong emphasis on causal inference
10+ years of experience applying causal inference and machine learning methods to real-world problems, with a demonstrated track record of measurable impact
Deep expertise in observational causal methods such as propensity score matching, Double Machine Learning, doubly robust estimation, instrumental variables, and difference-in-differences
Experience building reusable causal inference tools or platforms beyond one-off analyses
Proficiency with Spark, Ray, SQL, Python, and ML frameworks such as scikit-learn, XGBoost, and LightGBM
Experience with terabyte- or petabyte-scale datasets in distributed computing environments
Strong communication skills with the ability to translate econometric findings into clear business recommendations
Technology industry experience; connected TV, streaming, or advertising experience is a plus

Responsibilities

Design, build, and productionize a causal inference platform that standardizes how Roku measures the incremental impact of customer actions and business decisions
Research and implement causal estimation methods, including heterogeneous treatment effects, tailored to Roku's data and business questions
Build long-term outcome frameworks that enable impact projection from limited observation windows
Develop diagnostic and validation standards at scale to ensure credibility of causal estimates
Leverage AI to create counterfactual scenarios and build tools that help users run, understand, and act on causal estimates correctly
Work cross-functionally with Data Engineering, Product Management, and Core Analytics to translate business questions into well-defined causal problems and deploy production-ready solutions
Contribute to the technical vision of the Data Science team and the broader research agenda across causal inference, predictive modeling, and experimentation