Researcher, Frontier Robustness
OpenAI · San Francisco
Frontier labremote$295k–$500k
At a glance
Highlights
- Secure frontier AI by containing capable models under human control.
- Probe sandboxes and shared infra to find escapes and covert channels.
- Build and red-team defenses and monitors for agent reasoning.
What they offer
- Equity
Ideal candidate
Security-minded researcher who thrives on open-ended containment problems and turning experiments into deployed safeguards. Comfortable with uncertainty, hands-on with frontier models, and motivated to collaborate across research and infrastructure.
Requirements
- Hands-on experience conducting research with large language models, including designing and running experiments and evaluating model behavior
- Ability to turn an open-ended question into a clear hypothesis and build reproducible experiments at scale
- Ability to design rigorous evaluations and test findings across models, environments, and attack strategies
- Ability to write and debug code independently and build research tools and prototypes
- Practical security experience such as sandboxing, vulnerability research, adversarial testing, or detection and monitoring
- Ability to work with Security and infrastructure partners to develop threat models and turn research into deployed safeguards
Lab profile
About OpenAI
Builds safe frontier AI models and products like ChatGPT for broad benefit.
What they build
- ChatGPT
- OpenAI API
- Frontier AI models
Known for
- Frontier-scale distributed ML training
- Massive GPU fleet and data center operations
- Exabyte-scale data lakes and streaming
- AI alignment and safety research
Key facts
- Operates one of the world's largest GPU fleets
Posted · Last seen