Nuclear Materials Red-Team & AI Safety Analyst

Mercor; San Francisco (CA); Remote; USD 150,000 - 210,000; Full time

Job summary

Mercor seeks experts to assemble a panel evaluating frontier AI model misuse potential. You will craft prompts across benign, dual-use, and adversarial categories, assess model outputs, and prepare reference answers with clear rationale.

This role emphasizes writing-intensive work and precise analytic judgments. Applicants should demonstrate nuclear materials knowledge and the ability to articulate complex judgments in accessible language, with a track record in technical writing or expert

Qualifications

• Realistic nuclear materials knowledge is essential.
• Experience writing technical assessments or expert reports is a plus.
• Ability to distinguish routine professional questions from misuse signals.

Responsibilities

• Write challenging single-turn prompts in benign, dual-use, and adversarial categories.
• Evaluate model responses against a defined policy standard and judge correctness.
• Draft reference answers with technical reasoning for correct responses.

Skills

Nuclear forensics

Red-teaming

Technical writing

Policy reasoning

AI safety

Education

Background in nuclear science

Job description

Mercor seeks experts to assemble a panel evaluating frontier AI model misuse potential. You will craft prompts across benign, dual-use, and adversarial categories, assess model outputs, and prepare reference answers with clear rationale.

This role emphasizes writing-intensive work and precise analytic judgments. Applicants should demonstrate nuclear materials knowledge and the ability to articulate complex judgments in accessible language, with a track record in technical writing or expert