Frontier AI Safety Evaluator & Policy Auditor

Obsidian is seeking experienced AI Safety Practitioners to evaluate frontier AI models across grey-area topics, assess responses for safety, policy compliance, and quality, and provide structured feedback. You will review content on misinformation, political persuasion, self-harm, violence, and other high-risk areas, refine evaluation rubrics, and collaborate with researchers to improve model alignment.

Strong candidates have a 5+ years in AI safety or related fields, outstanding written

#J-18808-Ljbffr