Risk Analyst III

Title AI Red Teamer - AI Safety & Adversarial Testing Company Fortune 25 Company Location Austin, Texas Schedule Monday - Friday, Onsite Type Full-time, Ongoing W2, 12-Month Contract, New Role & Possible Extensions Pay Rate $50.25, with overtime pay Benefits 3 Weeks of PTO, Medical, Dental, Vision, 401(k), Life Insurance, Disability Coverage, Holidays Off but Unpaid Summary Our client is seeking creative, resilient, and highly motivated AI Red Teamers to support advanced AI safety initiatives. This role focuses on identifying vulnerabilities in cutting-edge AI systems through human-driven adversarial testing. Unlike automated testing, this position targets complex, high-risk scenarios requiring creativity, psychological insight, and out-of-distribution thinking. You will work across multiple modalities—including text, image, audio, and video—to uncover system weaknesses and evaluate real-world risk. This is a high-impact opportunity for individuals interested in AI safety, behavioral analysis, and adversarial problem-solving. The AI Red Teamer will collaborate closely with cross-functional teams to improve model safety, strengthen defenses, and support responsible AI deployment. Key Responsibilities Adversarial Testing & Attack Design Design and execute novel, multi-step adversarial test scenarios Apply techniques such as social engineering, roleplay, and emotional manipulation to evaluate model behavior Identify gaps in safeguards and surface unintended system outputs Vulnerability Assessment & Risk Evaluation Assess model outputs for real-world harm and risk exposure Prioritize vulnerabilities based on severity and user impact Support testing across multiple user types and scenarios Emerging Threat & Agentic Risk Testing Evaluate risks such as prompt injection, privilege escalation, and multi-system interactions Test edge cases involving complex or evolving AI behaviors Data Annotation & Reporting Document vulnerabilities and testing outcomes with clear, reproducible examples Annotate model responses and classify failure patterns Provide structured reports for engineering and safety teams Cross-Functional Collaboration Partner with researchers, engineers, and domain experts to improve system performance Contribute to the development of testing frameworks, taxonomies, and evaluation benchmarks Continuous Learning & Strategy Development Stay current with adversarial techniques, AI trends, and online behaviors Continuously refine testing strategies and methodologies Required Education & Experience Bachelor's degree preferred (or equivalent experience) Background in areas such as creative writing, psychology, mental health, humanities, or related fields preferred Required Skills & Qualifications Strong creative thinking and narrative-building ability Ability to identify psychological and behavioral vulnerabilities Experience analyzing complex scenarios and problem-solving in ambiguous environments Strong written and verbal communication skills Ability to document findings clearly for technical and non-technical audiences Adaptability across multiple content formats (text, image, audio, video) Comfort working in fast-paced and evolving environments Preferred Skills & Qualifications Experience in trust & safety, red teaming, adversarial testing, or risk analysis Familiarity with AI systems or generative AI tools Basic knowledge of prompt engineering or scripting techniques Understanding of AI safety or model evaluation concepts Important Notice This role may involve exposure to sensitive or objectionable content, including graphic or disturbing material. Candidates should be comfortable working in environments where such content may be encountered as part of testing and evaluation responsibilities. Support resources and wellness measures are provided. #LI-JH1