AI Red Team Researcher

This is a fully remote position, open to applicants in New York.

📋 Description

• Conduct ongoing adversarial evaluations of detection models across audio, video, image, and text mediums.

• Replicate innovative generative tools and methodologies, rigorously testing them against production models.

• Create realistic, comprehensive attack simulations that reflect actual bad-actor operations.

• Develop and sustain red team infrastructure, including testing frameworks, attack inventories, and adversarial datasets.

• Prioritize findings and deliver clear reproduction steps to AI scientists and engineers.

• Validate fixes through regression testing processes.

• Monitor generative AI tools, research, and threat actor tactics while evaluating potential weaponized techniques.

• Collaborate with Product teams to influence roadmap priorities.

• Assist Sales, Marketing, and Solutions Engineering with publishable research and threat assessments.

• Contribute to adversarial testing of APIs and integrations with Security.

• Help establish ethical guidelines and safe-handling protocols for synthetic media.

• Report directly to the VP of Product.

• Within 12 months, establish yourself as an internal authority on deepfake creation and evasion, expand adversarial testing infrastructure, influence product and customer risk discussions, and spearhead the testing of emerging threats.

⛳️ Requirements

• Over 4 years of experience in offensive security, adversarial machine learning, AI red teaming, or machine learning engineering with a strong adversarial emphasis.

• Practical experience in generating synthetic media, including voice cloning, text-to-speech, face swapping, lip-syncing, and diffusion-based image and video tools, in an offensive or research capacity.

• Proficient in Python programming.

• Solid understanding of machine learning concepts with expertise in PyTorch or similar frameworks.

• Capability to develop attack tools, automate testing pipelines, and analyze model failures.

• Familiarity with adversarial machine learning techniques, including evasion, adversarial perturbations, compression and post-processing attacks, and data poisoning.

• Proficient in measuring model robustness.

• Ability to communicate technical findings as prioritized, actionable insights for both technical and non-technical audiences.

• Background in media forensics, signal processing, speech/audio, or computer vision is a plus.

• Experience in traditional red teaming or penetration testing, relevant certifications such as OSCP, or familiarity with tools like Garak, PyRIT, or the Adversarial Robustness Toolbox is advantageous.

• Published research, conference presentations, CVEs, or contributions to open-source projects in adversarial ML, deepfakes, or AI security is a plus.

• Applicants must have authorization to work in the U.S.

• Willingness to relocate if currently not based in or near NYC for the headquarters-based option is required.

🏝️ Benefits

• Comprehensive healthcare plans with 100% premium coverage for employees and partial coverage for dependents.

• Dental and vision plans with 100% premium coverage for employees and their dependents.

• Short-term and long-term disability, as well as life insurance plans with 100% premium coverage for employees.

• Flexible Spending Account (FSA), Health Savings Account (HSA), and 401k programs.

• Equity compensation offered.

• 20 days of paid time off (PTO) annually.

• 12 weeks of parental leave.

• Budget allocated for learning and development.

• Monthly wellness benefits provided.

• Annual company-sponsored offsite event.

• Daily in-office lunch provided via UberEats for HQ employees.

• Commuter benefits available for HQ employees.

• Remote Fridays offered for HQ employees.

• Happy hours and other local events organized for HQ employees.

• Eligibility for an annual performance bonus.