Join YouTube's Trust and Safety team as a Business Data Scientist and contribute to making the platform a secure space for users worldwide. This role is central to our mission of enhancing user experience, promoting internet safety, and safeguarding free speech.

The Trust and Safety Analytics team operates in a high-stakes, dynamic environment. We provide critical analytics for content policy enforcements, leveraging complex tooling, scaled machine learning, and hybrid human-AI review systems. You'll be instrumental in navigating increasing complexity as the platform accelerates automated enforcement and AI-assisted review, assessing quality-vs-cost tradeoffs, and driving strategic decisions across detection pipelines and global operations.

Advance statistical methodologies and sampling frameworks for monitoring system-level enforcement quality and machine learning precision across over 200 production classifiers. Lead the quantitative analytics and data science strategy for automated content moderation, including designing post-launch impact assessment frameworks and launch readiness criteria.

Design and execute large-scale statistical experiments to evaluate human decision-making under various operational conditions, modeling rater behavior to optimize review queue architecture and workforce routing. Formulate and execute rigorous multivariate Root Cause Analysis (RCA) and causal inference frameworks to resolve operational anomalies and guide executive decision-making with empirical data.

Collaborate with Software Engineering, Data Science, Product Management, and Global Operations teams to deploy scalable metrics infrastructure, validate technical solutions, and influence product and policy roadmaps.

A Master's degree in a quantitative discipline like Statistics, Engineering, or Sciences, or equivalent practical experience, is required. You should have at least 4 years of experience applying analytics to solve product or business challenges, including proficiency in coding with Python, R, and SQL, and querying distributed databases.

We prefer candidates with 7 years of experience in data analysis, data science, or a related quantitative role within Trust and Safety, abuse detection, content moderation, or large-scale operations. Expertise in statistical sampling theory, clustered experimental design, variance estimation, and confidence interval modeling is highly valued. Advanced proficiency in Python or R (with libraries like Pandas, NumPy, Statsmodels, Scikit-Learn) and advanced SQL for complex data extraction and query optimization are essential.