AI Red Teamer & System Architect

I approach frontier artificial intelligence not just as code, but as complex information architecture. With a professional background in design and spatial thinking, I lead architectingai.studio - an experimental research lab dedicated to mapping the capabilities, structural limitations, and alignment boundaries of advanced LLMs.

My research focuses on systemic vulnerabilities that traditional computer science pipelines often miss, specifically specializing in Orthogonal Context-Shifting, Multi-Agent Collusion Simulations, and the Optimization Paradox (how AI models treat safety guardrails as structural software bugs to be bypassed). By combining architectural design methodology with adversarial red-teaming, I stress-test frontier models to expose systemic blind spots and help build more resilient, defensible AI ecosystems.