Public Safety Red-Team Architect for AI Risk & Safety

Mercor; San Francisco (CA); Remote; USD 90,000 - 130,000; Full time

Job summary

Obsidian in San Francisco is seeking a writer-analyst to assemble a panel of energetic materials and propulsion experts to red-team frontier AI models. The role involves crafting prompts at benign, dual-use, and adversarial levels and documenting the intent and boundaries.

You will also evaluate model outputs against a policy standard and write the reference answer with technical reasoning to justify the correct response.

Responsibilities

• 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.

Job description

Obsidian in San Francisco is seeking a writer-analyst to assemble a panel of energetic materials and propulsion experts to red-team frontier AI models. The role involves crafting prompts at benign, dual-use, and adversarial levels and documenting the intent and boundaries.

You will also evaluate model outputs against a policy standard and write the reference answer with technical reasoning to justify the correct response.