AI Safety Practitioner – Expert Evaluator

About the role

We are looking for an AI Safety Practitioner to serve as an Expert Evaluator for cutting‑edge AI research labs. The contractor will work remotely, reviewing AI‑generated outputs to ensure they meet safety, factual accuracy, and policy standards.

Key responsibilities

• Assess AI‑generated responses for safety, factual correctness, policy compliance, and overall quality.
• Review content that may involve misinformation, political persuasion, self‑harm, violence, cyber threats, biosecurity, and other high‑risk domains.
• Apply and refine evaluation rubrics for Reinforcement Learning from Human Feedback (RLHF), Supervised Fine‑Tuning (SFT), and AI safety benchmarking.
• Identify unsafe outputs, hallucinations, reasoning failures, and policy violations, providing structured feedback to improve model alignment.
• Collaborate with AI researchers and safety teams on ongoing evaluation initiatives.

Required profile

• Bachelor’s degree or higher in a relevant discipline (e.g., Computer Science, Psychology, Public Policy, Biology, Law).
• 5+ years of professional experience in AI safety, trust & safety, journalism, public policy, scientific research, or security.
• Excellent written English, strong critical‑thinking and analytical reasoning abilities.
• Proven ability to evaluate nuanced, policy‑sensitive scenarios consistently.

Required skills

• AI Safety evaluation methodologies.
• Experience with RLHF and SFT processes.
• Trust & Safety and content‑moderation frameworks.
• Development and application of evaluation rubrics.

What we offer

• Contract position paying $60–$70 per hour.
• Fully remote work with flexible scheduling.
• Opportunity to influence safety standards for leading AI research labs.