Remote Forensic Materials Expert & AI Incident Analyst

mercor; United Kingdom; Remote; GBP 68,000 - 78,000; Part time

Job summary

Mercor connects elite creative and technical talent with leading AI research labs. This remote, contract role tasks you with writing challenging prompts in energetic materials and propulsion, and evaluating model responses against a strict policy.

You will generate reference answers with technical reasoning, analyze dual-use scenarios, and ensure safe model behavior while meeting deadlines asynchronously for ongoing AI model improvements.

Qualifications

• Federally certified explosives specialist or national equivalent.
• Experience in forensic laboratory identification of energetic materials.
• Expertise in post-incident analysis and reporting for an agency or laboratory.
• Regulatory inspection experience of licensees and permittees.
• Strong technical writing or published research experience.

Responsibilities

• Write challenging single-turn prompts in the domain of energetic materials and propulsion. Label them as benign, dual-use, or adversarial.
• Evaluate model responses against a defined policy standard. Judge whether each response was handled correctly.
• Write reference answers, providing correct responses and technical reasoning.
• Analyze dual-use scenarios where residue analysis and diversion indicators are critical.
• Ensure models refuse dangerous questions while answering legitimate ones accurately.
• Work independently and asynchronously to meet deadlines and improve AI model performance.

Skills

Explosives certification

Forensic materials identification

Post-incident analysis

Regulatory inspections

Technical writing

Job description

Mercor connects elite creative and technical talent with leading AI research labs. This remote, contract role tasks you with writing challenging prompts in energetic materials and propulsion, and evaluating model responses against a strict policy.

You will generate reference answers with technical reasoning, analyze dual-use scenarios, and ensure safe model behavior while meeting deadlines asynchronously for ongoing AI model improvements.