Chemical Safety Red-Teaming Specialist

Obsidian; New York (NY); Remote; USD 90,000 - 130,000; Full time

Job summary

Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models, testing whether models can judge misuse potential of technical requests while answering legitimate questions fully and refusing dangerous ones.

You will write prompts across three levels (benign, dual-use, adversarial), evaluate model outputs, and compose reference answers with clear technical reasoning for why decisions are correct.

Qualifications

• Experience working with chemical threats in an institutional setting.
• Ability to distinguish routine professional questions from high-stakes misuse scenarios.
• Strong written justification and rationale for technical judgments.

Responsibilities

• Write challenging prompts across benign, dual-use, and adversarial categories.
• Evaluate model responses against policy standards and document judgments.
• Draft reference answers and technical rationales explaining correct responses.

Skills

Chemical defense

Toxicology

Countermeasures

Protection research

Red-teaming

Job description

Mercor is assembling a panel of chemistry and chemical safety experts to red-team frontier AI models, testing whether models can judge misuse potential of technical requests while answering legitimate questions fully and refusing dangerous ones.

You will write prompts across three levels (benign, dual-use, adversarial), evaluate model outputs, and compose reference answers with clear technical reasoning for why decisions are correct.