Become a vital part of Mercor’s mission as a Remote AI Safety Red Team Expert. Utilize your skills in adversarial probing and red teaming to enhance AI safety for customers.

At Mercor, we’re creating a specialized red team aimed at identifying vulnerabilities in AI systems. Your role will focus on testing conversational AI models and generating crucial human data to ensure robustness against biases and misinformation. Bring your solid curiosity and structured approach to support safer AI deployments across various projects.

Key Responsibilities: • Red team conversational AI systems with adversarial techniques • Generate high-quality human data through thorough analysis • Apply consistent structures using benchmarks and taxonomies • Document findings in reproducible reports and datasets

Requirements: • Native fluency in English and Danish • Prior experience in AI red teaming and adversarial work • Strong understanding of frameworks for structured testing • Excellent communication skills for diverse audiences • Adaptable in navigating various projects with clients

Your expertise will significantly boost AI safety, providing clients peace of mind in their systems. #J-18808-Ljbffr

📌 Remote AI Safety Red Team Expert (Toronto)
🏢 Mercor
📍 Toronto