At IBM, I assess AI-enabled products for prompt injection, jailbreaks, RAG vulnerabilities, sensitive data leakage, and excessive tool use. I developed LLM attack methodologies that became part of standard testing protocols.
Across 50+ product assessments, I identified 40+ high- and critical-severity vulnerabilities, including SQL injection, XSS, and IDOR/BOLA. I also led REST API and web application penetration tests, creating Burp Suite workflows mapped to the OWASP API Security Top 10.
I integrated security testing into GitLab CI pipelines and assessed Docker, AWS, and Kubernetes environments. I also developed Python automation for reconnaissance and vulnerability validation, and contributed API security playbooks used across 15+ engagements.
