Research Engineer, Privacy

About this role

OpenAI seeks a Research Engineer to develop and implement privacy-preserving techniques for AI systems. You'll design algorithms like differential privacy and federated learning, evaluate defenses against privacy attacks, and establish privacy standards across the ML lifecycle while collaborating with cross-functional teams.

What you'll do

• Design and prototype privacy-preserving ML algorithms deployable at scale
• Measure and strengthen model robustness against membership inference, model inversion, and data memorization attacks
• Develop internal libraries and evaluation suites for privacy techniques
• Investigate privacy–performance trade-offs of large models and publish findings
• Define privacy standards, threat models, and audit procedures for the ML lifecycle
• Collaborate with Security, Policy, Product, and Legal on regulatory compliance

What they're looking for

• Privacy-enhancing technologies (differential privacy, federated learning, secure aggregation)
• Deep learning frameworks (PyTorch, JAX)
• Privacy attack analysis and adversarial testing
• ML systems and production deployment
• Research publication and implementation
• Cross-disciplinary communication
• Threat modeling and security evaluation
• Technical documentation

Benefits

• Located in San Francisco with relocation assistance available