Cybersecurity and AI Red-Teaming AI Trainer

Project Overview

Become part of a progressive network of experts contributing to the cutting edge of artificial intelligence. In the role of AI Trainer focused on cybersecurity and red-teaming, you will actively assess and provide detailed insights on data to enhance large language model (LLM) effectiveness, ensuring future AI technologies maintain high standards of accuracy and trustworthy behavior.

This position is offered as a contract consultancy rather than a traditional employment role.

Location & Engagement

• Applicants must be based in the United States.
• The role is part-time and project-driven, allowing flexible timing and remote work.

Project Scope

Your work will support AI system red-teaming activities, focusing on identifying security vulnerabilities, testing resilience of AI models, crafting adversarial testing scenarios, and systematically documenting findings.

Contributions will support evaluations of advanced AI systems, emphasizing cybersecurity, safety, and reasoning capabilities.

Consultant Role Details

• The engagement is project-based with compensation determined per project, where hourly rates reflect expected effort.
• Consultants maintain autonomy over their schedules and tools and may engage with other clients or employers concurrently, respecting external policies.

Key Responsibilities

• Perform AI red-team assessments including jailbreak and prompt-injection tests.
• Create adversarial prompts and develop attack strategies.
• Assess AI outputs for robustness, safety compliance, and reliability.
• Spot and document AI model vulnerabilities, failure points, and weaknesses.
• Draft written rationales and comprehensive analysis reports.
• Engage in sequential, multi-turn AI testing procedures.
• Leverage cybersecurity and threat modeling expertise during AI evaluations.
• Collaborate closely with project teams by sharing documented insights and recommendations.

Project Deliverables

• Generation of high-quality adversarial test cases.
• Creation of detailed vulnerability and weakness reports.
• Provision of structured and clear evaluative judgments.
• Delivery of well-written justifications supporting findings.
• Strict adherence to defined project protocols and assessment frameworks.
• Reliable completion of tasks within established timelines.

Required Qualifications

• Proven professional background in cybersecurity, AI, or related analytical fields.
• Strong skills in adversarial thinking and complex problem solving.
• Experience in one or more areas such as AI red-teaming, offensive security, penetration testing, bug bounty hunting, capture the flag competitions, threat intelligence, application security, vulnerability research, AI safety, large language model evaluation, or machine learning and natural language processing engineering.
• Exceptional skills in written communication and documentation.
• Must be U.S.-based and qualified to work as an independent contractor.
• Preference for availability of at least 20 hours weekly.
• Ability to start work immediately or with minimal delay is favored.

Additional Information

Consultant compensation is project-based with estimated hourly rates from $60 to $70, depending on task complexity and time commitment.

Applicants should refer to the company's data security and privacy policies as outlined in the official Data Security and Privacy Notice.