Forensic & Toxicology Chemist - AI Safety Red Team

Neon seeks experienced writers to develop prompts at the boundary of safety for frontier AI models. You will craft labeled prompts (benign, dual-use, adversarial) and judge model outputs against a policy standard. Your findings will include a reference answer and the technical rationale behind it. This role emphasizes writing-intensive work, requiring prior technical writing, exposure to misuse scenarios, and the ability to clearly justify each judgment for non-specialists.