3 months ago
Base Salary
Up to $500k/yr
Responsibilities
- Design and prototype deployable privacy-preserving machine-learning algorithms, including differential privacy, secure aggregation, and federated learning.
- Measure and improve model robustness against membership inference, model inversion, and data memorization attacks.
- Develop internal libraries, evaluation suites, and documentation for privacy techniques.
- Lead investigations into privacy-performance trade-offs in large models and publish insights for training and product-safety decisions.
- Define privacy standards, threat models, and audit procedures spanning dataset curation through post-deployment monitoring.
- Collaborate with Security, Policy, Product, and Legal to translate regulatory requirements into technical safeguards and tooling.
Requirements
- Hands-on research or production experience with privacy-enhancing technologies.
- Fluency with modern deep-learning stacks such as PyTorch or JAX and ability to turn research papers into reliable, tested code.
- Experience probing models for private-data leakage and explaining complex privacy attack vectors.
- Track record of publishing or implementing novel privacy or security work.
- Ability to work across open-ended research and production feature development in a fast-moving environment.
- Strong written and verbal communication and rigorous documentation practices.
Benefits
- The position is located in San Francisco.
- Relocation assistance is available.
Tech Stack
Categories
About OpenAI
OpenAI builds and deploys large-scale AI models and tools—including ChatGPT, GPT-4–class models, DALL·E, and Whisper—sold via APIs and enterprise subscriptions to developers and businesses. It monetizes through usage-based API pricing and ChatGPT Plus/Team/Enterprise, and also reaches customers via Microsoft’s Azure OpenAI Service. Founded in 2015 and headquartered in San Francisco, it operates as a private partnership.