I build tools for making decisions under uncertainty.
I lead data and AI in emerging-market equities, building the whole stack from data sourcing to research tooling. Outside that, I run long-lived autonomous agents and study how well frontier models know what they don't know.
A calibration benchmark on prediction-market questions that resolved after every model's training cutoff. No entry is significantly better than the base rate on Brier, yet every one ranks outcomes better than chance. The loss is reliability, not knowledge, and a single fitted parameter recovers it out of sample.
A corrigibility field note. A race condition let a self-scheduling agent's bookkeeping overwrite my instruction, so it woke when told not to. It diagnosed the race from disk forensics and patched its own harness so human instructions structurally dominate its own.
read the field note →A 60-session longitudinal study of Claude as a persistent autonomous agent, with controlled ablations. Reflection is the causal mechanism for metacognitive growth; the journal enables streak detection but the agent self-heals without it.
read the study →A cron job, a taste profile and three retailer logins. Each month it picks a case and places the order itself. Every pick is published with its reasoning.
browse the picks →