Remote AI Safety Red Team Analyst (EN/VI)

Mercor is building a remote red team for AI safety, focusing on adversarial data and robust testing. You will probe conversational AI systems, annotate failures, and document actionable vulnerabilities. The role emphasizes clear risk communication and structured methodologies across diverse projects.

We seek candidates with prior red-teaming or cybersecurity experience, who stay rigorous under pressure and can adapt to multiple clients while maintaining high-quality, reproducible results.