Red Team Specialist (Austin, TX/Remote EST)

Our Customer’s mission is to give people the power to build community and bring the world closer together. Through their family of apps and services, they are building a different kind of company that connects billions of people around the world, gives them ways to share what matters most to them, and helps bring people closer together.

We are seeking a Red Team Specialist on a contract basis to help support our customers’ business needs. This role is on-site in Austin, TX - candidates in the Eastern time zone will also be considered.

Responsibilities:

• Design and execute creative, multi-turn adversarial attacks against AI systems across text, image, audio, and video modalities.
• Identify vulnerabilities in AI models through roleplay, emotional manipulation, social engineering, authority exploitation, and other adversarial techniques.
• Evaluate model outputs for real-world harm and risk severity using established risk taxonomies.
• Test for agentic AI vulnerabilities, including privilege escalation, indirect prompt injection, and scope creep within multi-authority systems.
• Generate high-quality human evaluation data through annotation, vulnerability classification, and adversarial test case development.
• Document reproducible model failures and communicate findings to engineering and safety teams.
• Collaborate cross-functionally with AI researchers, engineers, and domain experts to improve model safety and testing methodologies.
• Contribute to the refinement of red teaming taxonomies, benchmarks, and tooling infrastructure.
• Stay current with emerging adversarial techniques, AI safety research, internet subcultures, and evolving threat landscapes.
• Support the continuous improvement of AI safety testing processes and evaluation frameworks.
• Operate effectively within psychologically demanding environments involving exposure to sensitive and objectionable content.

Skills and Qualifications:

• 3+ years of relevant experience
• Background in creative writing, humanities, psychology, mental health counseling, special education, or related disciplines.
• Demonstrated ability to construct compelling narratives and identify linguistic or psychological vulnerabilities.
• Strong adversarial mindset with the ability to think creatively like an attacker.
• Ability to identify unexpected failure modes and vulnerabilities in AI systems.
• Strong adaptability across multiple content modalities, including text, image, audio, and video.
• Excellent written and verbal communication skills.
• Ability to clearly document and explain vulnerabilities to both technical and non-technical audiences.
• Strong analytical thinking and problem-solving capabilities.
• Ability to maintain resilience and professionalism while working with sensitive and potentially disturbing content.
• Strong collaboration skills within cross-functional technical and research environments.
• Ability to work independently within fast-paced and evolving AI safety environments.

Preferred Qualifications:

• Prior experience in professional red teaming, trust and safety, data annotation, or socio-technical risk analysis.
• Familiarity with large language models and generative AI platforms such as ChatGPT, Claude, or Gemini.
• Experience with prompt engineering, encoding techniques such as Base64 or ROT13, or scripting.
• Knowledge of AI safety concepts, including RLHF, alignment techniques, and model evaluation frameworks.

We offer a competitive salary range for this position. Most candidates who join our team are hired at the median of this range, ensuring fair and equitable compensation based on experience and qualifications.

Contractor benefits are available through our 3rd Party Employer of Record (Available upon completion of waiting period for eligible engagements)

* Health Benefits: Medical, Dental, Vision, 401k, FSA.

* Accrued PTO: Up to 15 days per 12 months on assignment

An Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, or protected veteran status and will not be discriminated against on the basis of disability.