Senior Causal Inference Data Scientist

Roku; San Jose (CA); Hybrid; USD 204,000 - 324,000; Full time

Benefits offered by this job

Health insurance

Equity awards

Life insurance

Disability benefits

Parental leave

Wellness benefits

Paid time off

Job summary

Roku is seeking a Senior Research Data Scientist in California to lead a causal inference platform measuring the true incremental impact of customer actions and features on long-term outcomes. You will partner with Product Growth to productionize scalable causal pipelines and provide high-confidence recommendations from observational data when experiments aren’t feasible.

Your work spans requirements, estimation approaches, and communicating results to senior leaders, with emphasis on

Qualifications

• PhD in Economics, Econometrics, Statistics, or related quantitative field with a causal focus.
• 10+ years applying causal inference and ML to real-world problems with impact.
• Expertise in observational causal methods and modern estimation techniques.

Responsibilities

• Design and productionize a causal inference platform for incremental impact measurement.
• Research and implement causal estimation methods for business questions.
• Build long-term outcome frameworks and scalable pipelines.
• Develop validation standards at scale for credibility of estimates.
• Collaborate with Engineering and Analytics to deploy solutions.
• Communicate findings to senior leadership and guide strategy.

Skills

Causal inference

Machine learning

Observational methods

Data analysis

Education

PhD in Economics/ Econometrics/ Statistics

Tools

Spark

Ray

SQL

Python

scikit-learn

XGBoost

LightGBM

Job description

Roku is seeking a Senior Research Data Scientist in California to lead a causal inference platform measuring the true incremental impact of customer actions and features on long-term outcomes. You will partner with Product Growth to productionize scalable causal pipelines and provide high-confidence recommendations from observational data when experiments aren’t feasible.

Your work spans requirements, estimation approaches, and communicating results to senior leaders, with emphasis on