AI security researcher focused on how agentic AI systems create attack surfaces that traditional AppSec tooling cannot see. My background is in securing critical systems across regulated industries: aerospace, banking, defense, and automotive.
More recently my work has centered on MCP and agentic AI: building lab-confirmed attack chains, publishing threat models, and mapping what breaks in production. That combination puts me in an unusual position: I understand how these systems are built, and I understand how they're broken.
What I focus on:
- Prompt injection & jailbreaks
- Data leakage through model outputs
- Agentic system risk & privilege escalation
- Supply chain & model integrity
- Threat modeling for agentic AI deployments
Where to find me:
- ✍️ aminrj.com — research & writing