Data Scientist, Meta Superintelligence Labs (Safety)

What you'll do

• Build telemetry and measurement frameworks to detect and monitor policy violations and emerging harms.
• Design scalable model evaluations and assess safeguards through offline and online testing, including human-in-the-loop and active learning.
• Run experiments and build signals, pipelines, and dashboards to evaluate mitigations and support safety interventions.

What they ask for

• Bachelor's degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
• 12+ years of experience in product data science, applied analytics, or quantitative research, including leading complex analytical workstreams as an individual contributor.
• 12+ years of experience with data querying and scripting languages and statistical or mathematical software for large-scale data analysis.
• Experience with advanced statistical methods, product experimentation, cross-functional stakeholder influence, and responsible AI practices.

Nice to have

• Advanced degree (Master's or PhD) in a quantitative field.
• Experience in frontier AI products or risks, or in Trust & Safety, Fraud/Risk/Security, model evaluations, threat modeling, actor telemetry, or human-in-the-loop review systems.

Pay

$210,000/year to $281,000/year USD base salary

Tools

SQL, Python, R

AI focus

Measure and mitigate AI model safety risks; scale model evaluations and safeguards; use AI tools to accelerate or redesign analytical workflows with responsible validation; develop AI skills such as prompt/context engineering and agent orchestration.