About this role
Help strengthen frontier AI systems by finding vulnerabilities before they can cause harm. In this remote, hourly role, you will use adversarial testing to probe model behavior across complex, high-risk, and ambiguous topics.
Key Responsibilities- Create challenging adversarial prompts to stress-test frontier AI models.
- Identify jailbreaks, unsafe behavior, hallucinations, policy failures, and other model weaknesses.
- Assess model robustness in misinformation, cybersecurity, biosecurity, fraud, political content, and other sensitive domains.
- Document vulnerabilities and contribute to safety benchmarks and red-teaming reports.
- Partner with AI researchers to improve model alignment, robustness, and safety.
- Bachelor''s degree or higher in Computer Science, Cybersecurity, Journalism, Communications, Psychology, Biology, Chemistry, Public Policy, or a related discipline.
- 5+ years of professional experience in AI safety, AI red teaming, trust and safety, cybersecurity, investigative journalism, life sciences, or a related field.
- Strong analytical reasoning, prompt-design, and written communication skills.
- Experience designing adversarial prompts or evaluating frontier AI systems.
- Experience with AI red teaming, RLHF, SFT, AI alignment, or trust and safety.
- Familiarity with jailbreak testing, prompt engineering, or adversarial evaluation methods.
- Expertise in one or more grey-area domains, such as cybersecurity, biosecurity, political content, misinformation, or scientific safety.
- Remote, hourly engagement.
- $70 to $84 per hour.
- Contribute to the safety of next-generation frontier AI models while working alongside AI researchers and safety teams on real-world safety challenges.