Radiation Safety Officer — AI Model Red-Teaming Lead

Mercor; Toronto; Vor Ort; CAD 90.000 - 150.000; Vollzeit

Zusammenfassung

Mercor is assembling a panel of radiological safety 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 prompts at three levels (benign, dual-use, adversarial), evaluate model responses, and craft the reference answer with detailed reasoning for how to respond correctly.

Qualifikationen

• Writing-intensive work requiring clear rationale for non-specialists.
• Prior technical writing, published research, or expert witness experience is a strong signal.
• Include a sample or link in your application or portfolio.

Aufgaben

• Write challenging single-turn prompts across three levels: benign, dual-use, adversarial.
• Evaluate responses against a defined policy standard and judge adherence.
• Write the reference answer and the technical reasoning explaining it.

Jobbeschreibung

Mercor is assembling a panel of radiological safety 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 prompts at three levels (benign, dual-use, adversarial), evaluate model responses, and craft the reference answer with detailed reasoning for how to respond correctly.