Red Team Engineer
Core
Break AI detection models using deepfake tools (voice clones, face swaps, injection attacks) to identify fraud and disinformation before adversaries do.
Role type
Senior IC adversarial ML engineer (AI red teaming)
Builds
Adversarial testing infrastructure, attack simulations, and repeatable test harnesses for deepfake detection models
Domain
Cybersecurity + Generative AI (audio, video, image, text)
Deliverable
production ML models
Required skills
adversarial ML techniques, Python, PyTorch, synthetic media generation (voice cloning, face swap, diffusion tools), model robustness measurement, attack automation
Preferred skills
media forensics, signal processing, computer vision, traditional red teaming, AI security tooling (Garak, PyRIT, ART), published research
Technologies
PyTorch, Garak, PyRIT, Adversarial Robustness Toolbox
Responsibilities
Run continuous adversarial testing across audio/video/image/text; build realistic end-to-end attack simulations; maintain test harnesses and adversarial datasets; close the loop on findings with AI/Data teams; track generative AI threat landscape; partner with Product on roadmap; support Sales/Marketing with threat briefings
Seniority
Senior, hands-on IC
