CBRN Safety RLHF Expert

About OpenTrain

OpenTrain AI is the hiring and contracting organization for this role. OpenTrain is the #1 platform for finding and building careers in AI training and data labeling, connecting skilled contributors with meaningful work that helps shape modern AI systems.

Creating an OpenTrain account is free, and candidates can apply in minutes.

About AI Safety and RLHF Work

AI training is the human side of building artificial intelligence. In reinforcement learning from human feedback, experienced reviewers assess model outputs, identify unsafe behavior, compare responses, and provide guidance that helps improve how AI systems respond.

This work combines subject-matter expertise with careful judgment. Your feedback can help models handle sensitive safety topics more accurately and responsibly.

• Review prompts and AI-generated responses
• Classify safety risks and evaluate model behavior
• Rank responses and provide expert reference feedback

The Role

As a CBRN Safety RLHF Expert, you will review text-based AI interactions and assess them through both a safety and subject-matter lens. The work requires professional experience in at least one relevant CBRN area, along with experience evaluating AI or language-model behavior.

This is an intermediate-level, part-time contractor opportunity paying $50 to $52 per hour. The expected commitment is less than 20 hours per week.

• Role type: Contractor and part time
• Pay: $50-$52 USD per hour
• Data type: Text
• Labeling focus: RLHF, red teaming, and evaluation rating
• Experience level: Intermediate
• Time requirement: Less than 20 hours per week

What You’ll Do

You will examine prompts and generated responses, determine whether content presents safety risks, and evaluate how effectively a model handles sensitive requests. You will also compare and rank responses and write expert reference feedback to support model improvement.

• Review prompts and AI-generated responses
• Classify CBRN-related safety risks
• Evaluate model behavior and response quality
• Rank multiple responses based on safety and effectiveness
• Provide expert reference feedback for AI training
• Apply red-teaming or content-safety judgment where relevant

Requirements

Applicants must have experience in at least one of the following domains: chemistry or biochemistry, radiology, nuclear, or explosives. Candidates must also have additional experience in AI or large language model evaluation, red teaming, AI safety, or content safety.

Fluent English proficiency is required.

• Professional experience in chemistry, biochemistry, radiology, nuclear, or explosives
• Additional experience in AI/LLM evaluation, red teaming, AI safety, or content safety
• Fluent English proficiency
• Intermediate-level experience suitable for independent evaluation work

Who Should Apply

This opportunity is designed for specialists who can combine CBRN knowledge with disciplined analysis of AI-generated content. It may suit professionals with safety, technical, research, evaluation, or red-teaming experience who want to contribute to the development of more reliable AI systems.

The role is available to eligible candidates in the United States, United Kingdom, Canada, Australia, New Zealand, Germany, France, Spain, Portugal, and Italy.

• CBRN specialists with experience assessing hazardous or sensitive content
• AI safety, LLM evaluation, or red-teaming professionals with relevant CBRN expertise
• Experts seeking flexible, part-time work in the growing AI training industry

Apply Through OpenTrain

AI training and data-labeling work gives specialists a direct role in improving cutting-edge AI. If you meet the CBRN and AI evaluation requirements, create your free OpenTrain account and apply for this contract opportunity.

• Review the requirements and confirm your relevant experience
• Create a free OpenTrain account
• Apply in minutes through OpenTrain