Mayukh Dey

I am an AI security specialist with hands-on experience in LLM red teaming, adversarial testing, and OSINT-driven investigations. I combine a technical background building production-grade tools with a policy-informed perspective from Political Science.

I have built and deployed multiple scanners and compilers, including an automated model and dataset scanner for public model hubs and a 17-module web vulnerability scanner that integrates LLM-powered payload generation and automated risk scoring. My tools emphasize reproducible detection, CVSS-aligned reporting, and practical exploit validation.

My technical toolkit spans Python, TypeScript, JavaScript, SQL, Bash, prompt engineering, LLM evaluation, and adversarial prompt design. I apply these skills to vulnerability assessment, WAF bypass techniques, threat intelligence analysis, and OWASP Top 10 coverage.

I hold multiple certifications in generative AI, AI security, governance, and MLOps, and I am seeking roles where I can advance AI trust, safety, and applied AI risk research through rigorous red teaming and secure tooling.