Frontier AI Safety Evaluator & Policy Reviewer

Obsidian; Paris; Sur place; EUR 70 000 - 90 000; Plein temps

Résumé du poste

Obsidian is seeking experienced AI Safety Practitioners to evaluate the safety, quality, and alignment of frontier AI models on grey-area topics. You will assess AI-generated responses, apply safety policies, and provide structured feedback to improve model behavior.

In this role you will review content across sensitive domains, refine evaluation rubrics for RLHF and SFT, and collaborate with AI researchers and safety teams to advance responsible AI deployment.

Qualifications

• Bachelor’s degree or higher in a related discipline
• 5+ years of professional experience in AI Safety or related field
• Excellent written English and analytical thinking
• Ability to evaluate nuanced policy-sensitive scenarios

Responsabilités

• Evaluate AI-generated responses for safety, accuracy, policy compliance, and overall quality.
• Review content involving misinformation, political persuasion, self-harm, violence, cyber, biosecurity, and other sensitive domains.
• Apply and refine evaluation rubrics for RLHF, SFT, and AI safety benchmarking.
• Identify unsafe outputs, hallucinations, reasoning failures, and policy violations.
• Provide structured feedback to improve model alignment and safety performance.
• Collaborate with AI researchers and safety teams on ongoing evaluation initiatives.

Connaissances

AI Safety experience

Analytical reasoning

Policy evaluation

Excellent written English

Formation

Bachelor’s degree or higher in a related discipline

Description du poste

Obsidian is seeking experienced AI Safety Practitioners to evaluate the safety, quality, and alignment of frontier AI models on grey-area topics. You will assess AI-generated responses, apply safety policies, and provide structured feedback to improve model behavior.

In this role you will review content across sensitive domains, refine evaluation rubrics for RLHF and SFT, and collaborate with AI researchers and safety teams to advance responsible AI deployment.