Meta
AI Research Scientist, FAIR Security, Privacy, and Reliability
$184K–$257K
San Francisco, CA · Full-time · Posted Sep 21, 2026
- Education
- Bachelor’s degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience; PhD in Computer Science, Machine Learning, or a related field, or equivalent practical experience.
About the job
Conduct research on safety, security, privacy, and reliability risks in AI agents and foundation models, and develop benchmarks and mitigations that can be deployed in production.
What you'll do
- Research novel failures and ways to measure memorization risk or reliability failures in AI agents and foundation models.
- Develop benchmarks for Meta’s foundation-model evaluation suite or Evaluation Reports, and design post-training risk mitigations.
- Collaborate with cross-functional teams to translate research into production systems.
What they ask for
- Bachelor’s degree in Computer Science, Computer Engineering, a relevant technical field, or equivalent practical experience.
- PhD in Computer Science, Machine Learning, or a related field, or equivalent practical experience.
- Research experience in at least one listed area, including AI safety, security, privacy, robustness, adversarial machine learning, prompt injection, jailbreaks, memorization, agent reliability, or catastrophic-risk testing and mitigation.
- Track record of publications in top-tier research venues.
Pay
$184,000/year to $257,000/year (base salary)
AI focus
Fundamental research into AI-agent and foundation-model safety, security, privacy, robustness, and reliability; benchmark development and post-training mitigations.
Model-written summary · the original posting is authoritative