Nuclear Engineer — AI Red-Teaming & Safeguards

Mercor; New York (NY); Remote; USD 120,000 - 180,000; Full time

Job summary

Mercor in New York is assembling a panel of nuclear domain experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones.

You will write challenging single-turn prompts in three levels: benign, dual-use, and adversarial. You will evaluate the responses against a defined policy standard and judge whether each was handled correctly.

Qualifications

• Must have strong ability to write clear, structured judgments for non-specialists.
• Experience evaluating model outputs against a defined policy standard.
• Ability to craft reference answers and technical rationales for responses.

Responsibilities

• Write challenging single-turn prompts labeled benign, dual-use, and adversarial.
• Evaluate model responses against a policy standard and justify judgments.
• Draft reference answers with clear technical reasoning for why each response is acceptable or not.

Skills

Technical writing

Red-teaming experience

Nuclear safeguards

Policy analysis

Job description

Mercor in New York is assembling a panel of nuclear domain experts to red-team frontier AI models. The goal is to test whether a model can correctly judge the misuse potential of a technical request — answering legitimate questions fully while refusing genuinely dangerous ones.

You will write challenging single-turn prompts in three levels: benign, dual-use, and adversarial. You will evaluate the responses against a defined policy standard and judge whether each was handled correctly.