Researcher, Safety & Privacy
Core
Design and build privacy-preserving safety systems for frontier AI models to detect and mitigate harms like jailbreaks and weaponization instructions without exposing user data.
Role type
Researcher, Privacy-Preserving Safety
Builds
Automated privacy-preserving safety systems and auditable frameworks for harm detection
Domain
AI Safety, Cryptography, Security, Privacy
Deliverable
production ML models
Required skills
Privacy-preserving computation (MPC, secure enclaves, differential privacy), Security and adversarial systems, Machine learning safety or alignment, Algorithmic auditing, Secure system design
Preferred skills
AI safety, jailbreak detection, model alignment, Privacy-preserving machine learning techniques
Technologies
MPC, secure enclaves, differential privacy
Responsibilities
Design and implement privacy-first architectures for detecting harmful model behaviors, Build frameworks for auditable private identification of high-risk content, Develop strict, auditable mechanisms triggered by harm signals, Drive development of automated safety systems that preserve privacy
Seniority
Senior, hands-on IC