Frontier AI Safety Red Team Specialist

Il y a 5 heures

Arrondissement of BrusselsCapital, Brussels, Belgique Mercor Temps plein
pMercor is seeking experienced AI Safety Red Teamers to identify vulnerabilities in frontier AI systems through adversarial testing. You will design challenging prompts, uncover model weaknesses, and evaluate AI behavior across complex, high-risk topics. /ppResponsibilities include designing adversarial prompts, identifying jailbreaks and policy failures, and evaluating model robustness across sensitive domains. /p