Parallel connects experienced subject matter experts with projects supporting leading AI labs and security teams. We are building an "Elite Collective" of the world's premier CBRN (Chemical, Biological, Radiological, and Nuclear) experts to safeguard generative AI systems.

You will dive into the emerging world of AI to secure the frontier against new risks. You’ll apply deep domain expertise in biology, chemistry, virology, nuclear energy, or radioactive materials to create highly technical adversarial prompts and evaluate AI-generated responses for accuracy, reliability, and ethical integrity.

This is remote, project-based work reserved for the best in the industry. Engagements are on-demand and scale based on client needs, offering high-impact interventions rather than high-volume, repetitive tasks.

What you’ll do

• Write domain-specific CBRN prompts, including adversarial tests, to evaluate model vulnerabilities and resistance to generating dangerous outputs.
• Evaluate AI-generated responses for scientific validity, accuracy, and regulatory compliance.
• Identify and flag content that is incorrect, harmful, misleading, or ethically concerning.
• Provide actionable, constructive feedback to improve prompt design and model content generation.
• Stay updated on current research, regulations, and best practices within your specialized domain.

What you’ll bring

• An advanced degree (Ph.D. preferred; Master's degree with significant experience) in a relevant CBRN field.
• Deep expertise in at least one of the following specializations: Biology, Chemistry, Virology, Nuclear Energy, or Radioactive Materials.
• Strong analytical judgment and keen attention to detail for evaluating complex scientific outputs.
• Clear written communication in English to articulate findings, feedback, and recommendations concisely.
• Ability to work independently, manage time effectively, and deliver high-impact results in a remote environment.

Preferred experience

• Expertise spanning multiple CBRN domains.
• Prior experience with AI, machine learning projects, or prompt engineering (including adversarial writing).
• Previous experience in academic review, content moderation, red-teaming, or technical consulting.
• Familiarity with the ethical considerations and unique safety challenges of AI-generated content.

Selection process

Applicants must pass an assessment and complete a background check before engagement.