Researcher, Safety Training, National Security
AI · 56 roles open in software and data
- Location: San Francisco
- Pay: $380k–$500k a year
- Experience: 4+ yrs
- Posted: Posted
About the role
The team researches methods to train AI models for safe and reliable behavior, particularly for national security applications. The researcher develops and evaluates these training techniques, focusing on policy adherence and robustness against misuse.
Our summary of the posting; the original is on the OpenAI careers page.
What the posting requires
- Required
- Security clearance
- RLHF
- Deep learning
- Adversarial training
- Research
- Robustness
- Engineering
- Computer science
- Machine learning
- Preferred
- Safety training
- Reinforcement learning
- Adversarial robustness
- Evaluations
- Model failure modes
- Improve training
These tags are our reading of the posting. They can miss something; the original posting is the reference.
Do you qualify?
Each requirement checked against your resume, in about a minute. No account. Your resume is deleted when the verdict appears.