Roblox is seeking a Principal Data Scientist for Safety to help define, measure, and improve detection and mitigation of harmful user-generated content and violative behaviors on the platform. In an onsite San Mateo, CA role, you will combine data science, statistics, and causal inference to generate authoritative safety measurement, build high-quality ground truth datasets, and translate evidence into practical guidance across Product, Engineering, and policy-focused partners.
What you’ll do
- Establish and monitor robust prevalence measurement methodologies that quantify the overall level of harmful content and behavior, serving as the organization’s source of truth for safety goals.
- Develop and validate ground truth datasets for harmful UGC and violative on-platform behaviors, ensuring quality and alignment with current platform policies.
- Perform exploratory analysis to deepen understanding of violations and emerging threat landscapes, providing data-backed recommendations on where Product and Engineering should invest in detection and moderation resources.
- Design, implement, and analyze experiments for new moderation and safety enforcement features, sharing results with Product and Engineering partners to support development decisions.
- Partner closely with Policy, Compliance, and Legal teams to ensure strategic alignment around safety initiatives.
- Apply causal inference methodologies to measure effectiveness and potential systemic impacts of safety initiatives on player experience and platform integrity.
- Communicate strategic insights to leadership and cross-functional partners, translating findings on prevalence, ground truth quality, and effectiveness into actionable plans.
- Work with ML and Data Engineering teams so model development, reporting, and detection systems rely on statistically sound ground truth and measurement frameworks.
Requirements
- 10+ years of industry experience in data science, economics, analytics, or machine learning engineering.
- 7+ years using scripting languages such as Python and R, plus big data query or processing tools including SQL, Hive, Spark, and Airflow.
- Knowledge of machine learning and deep learning via formal training or industry experience.
- Ability to use creative first-principles reasoning to tackle ambiguous problems.
- Experience building large-scale safety or moderation systems, with experience in content platforms, specifically user-generated content.
- Advanced degree and/or PhD in Statistics, Computer Science, Physics, Applied Math, Economics, or another related quantitative field.
Tools and technologies
- Python, R, SQL
- Hive, Spark, Airflow
Salary and work location
The annual salary range for this role is $274,000 to $373,680 USD.
Location: San Mateo, CA (onsite). Onsite attendance is Tuesday, Wednesday, and Thursday, with optional presence on Monday and Friday unless otherwise noted.
Work authorization notice (US-based roles)
For US based roles only: the company may not be able to employ candidates for this role who have United States work authorization related to certain U.S. visa categories, or support future H-1B sponsorship at this time.
