Persona_Identity_Report

System Context.

“Bridging the gap between high-dimensional theoretical research and secure, scalable AI implementations.”

Technical_Stack

  • Architecture LLMs / RAG
  • Security Threat Modeling
  • Core_Langs Python / C++
  • Frameworks PyTorch / Strands

The Journey

I'm the Co-Founder & CTO of Armorer Labs (YC F26), where we build continuous security verification for AI agents. I lead our AI research, evaluation, and engineering. Before Armorer, I led AI security science at Amazon Web Services (AWS), building and evaluating AI systems for security.

“I believe that the next decade of software will be defined not by the code we write, but by the constraints we place on the models that write it for us.”

My academic foundation was built at Columbia University, where I completed an M.S. in Applied Analytics (GPA 4.1/4.0), and Università Bocconi, where I earned my B.S. in Economics and Management. This dual background allows me to communicate technical nuances to stakeholders while maintaining the mathematical rigor required for state-of-the-art AI development.

My research focuses on AI security and model internals. I'm the lead author of ThreatForest, an open-source agent that turns source code into attack trees (presented at Black Hat USA 2026), and a co-author of SIR-Bench, a benchmark for evaluating AI security investigations. I also build open-source projects like InfluencerPy and share deep dives into LLM interpretability on Medium.

Explore Weights →