Offensive Security Analyst, AI Specialist

Vanguard

full-time

Posted on: 8/27/2025

Origin: • 🇺🇸 United States • Pennsylvania

✨ AI Apply

Mid-LevelSenior

AWSAzureCloudGoogle Cloud PlatformPythonPyTorchScikit-LearnTensorflow

About the role

Design and conduct Red Team operations focused on AI/ML systems, including adversarial input testing, model inversion, poisoning, and evasion attacks
Emulate threat actors targeting AI infrastructure (e.g., model hosting platforms, training data lakes, inference APIs)
Develop custom tooling and payloads for AI-specific attack scenarios, including adversarial examples and synthetic data manipulation
Collaborate with data scientists and ML engineers to assess model robustness, fairness, and explainability under adversarial conditions
Lead Purple Team exercises to validate detection and response capabilities for AI-related threats
Produce detailed reports and executive briefings outlining risks, attack paths, and remediation strategies
Contribute to threat modeling and detection engineering for AI systems
Stay current on emerging AI threats, adversarial ML research, and regulatory implications
Mentor junior team members and foster a culture of offensive innovation

Proven experience in Red Team operations, adversary emulation, or advanced penetration testing
Deep understanding of AI/ML concepts including supervised/unsupervised learning, NLP, computer vision, and reinforcement learning
Experience with adversarial ML techniques (e.g., FGSM, PGD, DeepFool, model extraction)
Strong proficiency in Python and ML frameworks (e.g., TensorFlow, PyTorch, Scikit-learn)
Familiarity with cloud platforms (AWS, Azure, GCP) and their AI services
Knowledge of MITRE ATLAS, MITRE ATT&CK, and NIST AI RMF
Experience with offensive tooling and scripting for automation and exploit development
Excellent communication skills for technical and executive audiences
Experience mentoring junior team members
Sponsorship not offered for this position