Description

You will perform adversarial testing on GenAI systems to identify and remediate security vulnerabilities.

This role is hybrid.

Responsibilities

• Execute hands-on adversarial testing including prompt injection, jailbreaks, input-output evaluations, and data-exfiltration tests.
• Develop actionable remediation strategies based on testing findings.
• Build reusable adversarial test repositories, scripts, and automation harnesses.
• Apply cybersecurity red-teaming and penetration testing methodologies to ML/GenAI environments.

Required Skills

• 5+ years of experience in cybersecurity, red-teaming, or penetration testing.
• Deep understanding of GenAI fundamentals including LLMs, embeddings, and diffusion models.
• Practical experience with adversarial ML techniques such as model extraction, poisoning, and prompt injection.
• Proficiency in Python and scripting for automation and test harness development.
• Experience with AI/MLOps platforms and integration frameworks like Azure AI, AWS SageMaker, OpenAI API, Hugging Face, or LangChain.
• Familiarity with AI security frameworks such as NIST AI RMF, MITRE ATLAS, or OWASP Top 10 for LLMs.
• Knowledge of enterprise-scale AI/ML integration.

Preferred Skills

• Exposure to AI governance and risk management, including model risk and policy alignment.
• Experience with SIEM/SOAR and threat-intel integration and monitoring.

Key Skills

Education

Any Gradute