A working catalogue of
decision systems.

VELA turns a voice request for care into a priced, evidence backed, consent approved path by combining local AI, real hospital transparency data, insurance context, and auditable action.

An active research prototype exploring whether an AI opponent can feel co-present at a real chessboard. The interface holds conversation, spoken moves, and status. It holds no digital board. Live camera move recognition is still an open problem.

A conversational research prototype that organizes inquiry through four observation lenses — Dialectic, Socratic, Reflection, and Transfer — preserving assumptions, tensions, uncertainty, and the next cognitive move rather than producing an immediate answer.

A source-grounded civic intelligence prototype that turns a public-meeting record into inspectable decisions, commitments, deadlines, geographic context, and resident next steps.

A public research prototype in which five specialized agents — planner, researcher, skeptic, analyst, and replicator — carry a visitor's question through a six-stage, fully inspectable research run. The public release is an interactive simulation, not live empirical research.

A reproducible empirical research platform that integrates institutional 13F holdings, official information events, and retail-attention signals to surface transparent, non-causal directed-herding candidates for behavioral finance research.

A multi-agent reasoning system that turns an event or policy decision into an inspectable graph of cascading consequences across interconnected domains.

An evolving AI literacy research notebook: from a fixed confirmation-bias prototype to a prompt-driven laboratory where people manipulate agent context and learn intuitively how evidence shapes conclusions.

Automated inventory planning and decision-support system. Surfaces only the SKUs that need a human decision today.

A six-version record: an SAP and planning reporting foundation (In Development) that became the Financial Variance Command Center — a synthetic six-agent investigation room where every variance carries a theory, a trace, and a human disposition.

An AI-mediated thinking environment that improves human judgment through Socratic reflection, metacognitive questioning, and longitudinal analysis of the decision-making process.