Nuclear Safeguards Red-Teaming Specialist

Mercor; Paris; À distance; EUR 70 000 - 110 000; Plein temps

Résumé du poste

Mercor is assembling a panel of nuclear materials and safeguards experts to red-team frontier AI models, testing whether a model can judge misuse potential while refusing dangerous requests.

You will write prompts at three levels, evaluate model responses against a policy standard, and craft reference answers with technical reasoning. This is writing-intensive, with emphasis on clear justification for non-specialists.

Qualifications

• Examples of relevant backgrounds include material control and accountability (MC&A) programme design, operation or assessment; physical protection systems; vulnerability assessment for nuclear facilities; insider threat and human reliability; nuclear security policy, training or international assistance.

Responsabilités

• Write challenging single-turn prompts in your domain, labelled across three levels: benign, dual-use, and adversarial.
• Evaluate the responses that come back against a defined policy standard, and judge whether each was handled correctly.
• Write the reference answer — what a correct response looks like, and the technical reasoning for why.

Connaissances

MC&A expertise

Physical protection systems

Vulnerability assessment

Insider threat awareness

Nuclear security policy

Description du poste

Mercor is assembling a panel of nuclear materials and safeguards experts to red-team frontier AI models, testing whether a model can judge misuse potential while refusing dangerous requests.

You will write prompts at three levels, evaluate model responses against a policy standard, and craft reference answers with technical reasoning. This is writing-intensive, with emphasis on clear justification for non-specialists.