Projects

Research

CogVuln

A cyber attack modeled as a game between two agents with minds. Red, the attacker, works a reconnaissance-to-compromise kill chain, but can be endowed with human decision biases like base-rate neglect, loss aversion, and the sunk-cost fallacy. Blue, the defender, watches Red act and performs Bayesian inference over hypotheses about which bias is distorting Red’s choices, turning an attacker’s psychology into a defensive signal.

CogVuln · Python · PsychSim · Bayesian inference

Writing

Rethinking Triage: Designing Systems for Routing, Context, and Ticket Quality

Taking the manual coordination out of support triage at Advisor360°. An LLM assigns each incoming ticket to the team that owns the problem, embeddings and cosine similarity surface the past tickets that already solved it, and a validation step holds every submission to what happened, where, and when. Enterprise Shared Services went from absorbing 20% of ticket volume to 5%, and time to first assignment became far less variable.

Advisor360° · April 2026 · LLM routing · embeddings

Smaller projects and experiments live on GitHub ↗.