Principal Data Scientist, Safety & Moderation

Roblox Corporation; San Mateo, Northern (CA, KY); Hybrid; USD 274,000 - 374,000; Full time

Job summary

Roblox Corporation in San Mateo, CA, is seeking a senior data scientist for Data Science & Analytics to lead safety and moderation initiatives at scale.

You will develop prevalence measurement methodologies, create ground truth datasets for harmful UGC, and apply causal inference to assess safety initiatives. Collaborate with Product, ML, and Policy teams to drive impactful decisions.

Qualifications

• 10+ years of industry experience in data science, economics, analytics, or machine learning engineering.
• 7+ years of experience using Python or R and big data tools such as SQL, Hive, Spark, and Airflow.
• Experience developing large-scale safety or moderation systems for user-generated content.
• Advanced Degree and/or PhD in Statistics, CS, Physics, Applied Math, Economics, or related quantitative field.

Responsibilities

• Establish and monitor robust prevalence measurement methodologies for harmful content and behavior on the platform.
• Develop and validate ground truth datasets for harmful UGC and violative on-platform behaviors.
• Deepen understanding of violations via exploratory analysis and provide data-backed recommendations.
• Design, implement, and analyze experiments for moderation and safety features with cross-functional partners.
• Apply causal inference to measure effectiveness of safety initiatives on user experience and platform integrity.
• Communicate insights and recommendations to leadership and cross-functional teams.
• Partner with ML and Data Engineering to ensure models are built on solid ground truth and measurement frameworks.

Skills

Python

R

SQL

Hive

Spark

Airflow

ML

Deep Learning

Education

Advanced Degree/PhD in a quantitative field

Job description

Roblox Corporation in San Mateo, CA, is seeking a senior data scientist for Data Science & Analytics to lead safety and moderation initiatives at scale.

You will develop prevalence measurement methodologies, create ground truth datasets for harmful UGC, and apply causal inference to assess safety initiatives. Collaborate with Product, ML, and Policy teams to drive impactful decisions.