Business Data Scientist, YouTube Trust and Safety

Written by the T&S Network from Google's posting.

Analytics for YouTube's content policy enforcement, covering measurement of enforcement quality and classifier precision across 200+ production systems. The team tracks safety metrics and creator appeals and reinstatements while automated enforcement and AI-assisted human review expand. The role designs experiments on rater decisions and workforce routing.

What you'd do

• Maintain statistical methods and sampling frameworks that monitor quality across 200+ production classifiers.
• Set launch readiness criteria and assess impact of automated policy enforcement systems after launch.
• Design large-scale experiments on how human reviewers decide under different operating conditions.
• Model rater behavior to inform review queue architecture and workforce routing.
• Run root cause analyses and causal inference on operational anomalies for executive decision-making.
• Build metrics infrastructure with Engineering, Data Science, Product, and Global Operations teams.

What it asks for

• Master's degree in a quantitative field such as Statistics, Engineering, or Sciences, or equivalent practical experience.
• 4 years using analytics, coding (e.g., Python, R, SQL), or statistical analysis on product or business problems.
• 4 years querying large quantitative datasets in SQL across distributed databases.
• 6 years using analytics and coding on product or business problems (preferred).
• Strong grounding in sampling methods, clustered experiment design, variance estimation, and confidence intervals (preferred).
• Advanced proficiency in Python or R with libraries such as Pandas, NumPy, Statsmodels, or Scikit-Learn (preferred).

Our best summary of the posting, and it can be wrong. The employer's posting is what counts: read it in full on Googlenorth_east before you apply. If we got something wrong, report this listing.