Frontier AI Safety Evaluator

Mercor in Madrid is seeking experienced AI Safety Practitioners to evaluate frontier AI models on safety, quality, and alignment across sensitive, policy-driven topics. You will assess AI-generated responses, apply safety policies, and provide structured feedback to improve model behavior. You will refine evaluation rubrics for RLHF and SFT, identify safety violations and hallucinations, and collaborate with researchers and safety teams on ongoing evaluation initiatives. J-18808-Ljbffr