AI/ML Security Specialist

Job Responsibilities

• Conduct Operational Testing and Evaluation (OT&E) of AI/ML applications, models, and supporting infrastructure to identify security vulnerabilities, operational risks, and potential attack vectors.
• Perform adversarial AI/ML security assessments, including model evasion, data poisoning, model extraction, adversarial manipulation, and model inversion testing.
• Evaluate existing cybersecurity defenses and identify weaknesses that could be exploited through AI-enabled attacks, recommending appropriate mitigation strategies and security controls.
• Design, implement, and evaluate defensive security measures to protect AI/ML models, training data, inference pipelines, and deployed applications against emerging threats.
• Develop detailed technical assessment reports documenting vulnerabilities, testing methodologies, risk severity, and actionable remediation recommendations.
• Create and maintain Standard Operating Procedures (SOPs), security testing plans, threat models, and repeatable methodologies for AI/ML security evaluations.
• Collaborate with cybersecurity engineers, AI/ML developers, technical leads, and federal stakeholders to integrate security best practices throughout the AI/ML development and deployment lifecycle.
• Ensure AI/ML security assessments, testing procedures, and defensive implementations align with DHS, CBP, NIST, and applicable federal cybersecurity policies and standards.

Job Qualifications

• Demonstrated experience in cybersecurity engineering, security assessments, or penetration testing, with a strong focus on AI/ML security, adversarial machine learning, and model evaluation.
• Strong understanding of AI/ML architectures, machine learning algorithms, large language models (LLMs), model training pipelines, and secure deployment practices.
• Hands-on experience identifying and evaluating AI-specific security vulnerabilities, including prompt injection, adversarial examples, data poisoning, model extraction, model inversion, and evasion attacks.
• Experience conducting penetration testing, vulnerability assessments, red teaming, or Operational Testing and Evaluation (OT&E) of software applications, AI/ML models, or supporting infrastructure.
• Proficiency with Python and familiarity with AI/ML frameworks such as PyTorch, TensorFlow, or scikit-learn, along with security testing tools and adversarial assessment methodologies.
• Knowledge of federal cybersecurity requirements and security frameworks, including NIST AI Risk Management Framework (AI RMF), NIST SP 800-53, NIST Cybersecurity Framework, and applicable DHS/CBP security policies.
• Experience developing technical assessment reports, threat models, security testing methodologies, remediation recommendations, and Standard Operating Procedures (SOPs).
• Strong analytical, problem-solving, and written and verbal communication skills, with the ability to present complex technical findings and security recommendations to federal stakeholders, engineering teams, and technical leadership.