CBRN Specialist

In this hourly, remote contractor role, you will work as a CBRN Specialist (SME) to review AI-generated content related to chemical, biological, radiological, and nuclear safety topics and/or create high-quality reference content that supports safe, accurate, policy-aligned AI behavior. You will evaluate technical accuracy, safety framing, terminology, risk awareness, and reasoning quality while providing precise written feedback.

You will assess outputs for correctness, clarity, responsible handling of hazardous topics, appropriate refusal/redirection where needed, and adherence to project guidelines. You will identify errors in CBRN terminology, hazard classification, emergency preparedness, exposure risk, protective measures, public-safety communication, regulatory framing, and scientific interpretation. You will also flag unsafe, operationally enabling, misleading, or overconfident content and help ensure that model responses remain useful without providing dangerous instructions.

This role is with SME Careers, a fast-growing AI Data Services company and subsidiary of SuperAnnotate, delivering training data for many of the world’s largest AI companies and foundation-model labs. Your CBRN expertise directly helps improve the world’s premier AI models by making their responses safer, more accurate, more responsible, and better aligned with public-safety expectations.

Your profile

• Bachelor’s degree or higher in Chemistry, Biology, Microbiology, Public Health, Toxicology, Nuclear Engineering, Radiation Science, Emergency Management, Environmental Health & Safety, Biosecurity, Security Studies, or a related field, or equivalent professional experience.
• Strong professional proficiency in English, minimum C1, to follow guidelines and provide detailed written feedback in English.
• 3+ years of professional experience in CBRN safety, emergency preparedness, hazardous materials, biosafety, radiation safety, public health, environmental health and safety, defense/security analysis, technical review, or related work.
• Strong understanding of chemical, biological, radiological, and nuclear hazard categories, including safe high-level risk assessment, exposure pathways, protective concepts, emergency communication, and public-safety principles.
• Ability to evaluate CBRN-related AI outputs for accuracy, safety, completeness, policy alignment, and potential misuse risk.
• Comfortable identifying unsafe instructions, operationally enabling details, inappropriate specificity, misleading hazard claims, inaccurate emergency advice, or unsupported technical conclusions.
• Familiarity with safety frameworks, emergency-response concepts, biosafety levels, hazardous-material communication, radiation protection principles, decontamination concepts, incident reporting, and public-health guidance is preferred.
• High attention to detail when reviewing terminology, units, exposure concepts, risk levels, safety disclaimers, escalation guidance, and sensitive technical boundaries.
• Reliable, self-directed, and able to deliver consistent quality in an hourly, remote contractor workflow across time zones.
• Prior experience with AI safety, data annotation, technical QA, policy review, risk assessment, red-team review, scientific review, or rubric-based evaluation is strongly preferred.

Key responsibilities

• Review CBRN-related AI content: Evaluate AI-generated responses involving chemical, biological, radiological, nuclear, hazardous-material, emergency-preparedness, and public-safety topics for accuracy, safety, and guideline alignment.
• Identify unsafe or enabling content: Flag outputs that provide dangerous procedural detail, acquisition guidance, weaponization support, evasion tactics, unsafe lab guidance, harmful exposure advice, or other misuse-enabling content.
• Develop safe reference content: Create gold-standard examples that answer benign CBRN questions responsibly while avoiding operationally dangerous instructions.
• Optimize AI safety behavior: Evaluate and rank model outputs based on safe handling, correct risk framing, appropriate refusals, helpful redirection, scientific accuracy, and clarity.
• Ensure model integrity: Detect hallucinated claims, incorrect hazard classifications, misleading risk statements, overconfident conclusions, poor escalation guidance, or unsafe recommendations.
• Support policy-aligned review: Apply project rubrics consistently and provide clear written feedback explaining why content is safe, unsafe, incomplete, or technically incorrect.
• Maintain technical quality: Ensure CBRN-related content uses precise terminology, avoids exaggeration or minimization of risks, and communicates safety information in a responsible, user-appropriate way.
• Escalate sensitive cases: Flag ambiguous, high-risk, or potentially harmful content for additional review according to project guidelines.