Researcher / Engineer
Reliable decisions in uncertain worlds.
I study intelligent systems that must reason with partial information, strategic behavior, and unreliable evidence.
Safe Observation Capacity
A safe method for learning exploitable opponent behavior when folds censor private-card observations.
CFR Without Unbiasedness
Deterministic convergence guarantees and exploitability certificates for persistent partial public-chance evaluation.
LLM Exploitability Audits
Exact best-response audits reveal worst-case vulnerabilities that selected-opponent evaluations can miss.
02 / Point of view
Reliability is a property to test, not a label to claim.
- 01
Observe
Identify which information changes a decision and what that information costs.
- 02
Reason
Expose assumptions, strategic incentives, and the strongest available counterexample.
- 03
Certify
State what the system can guarantee and where that guarantee stops.
CFR without unbiasedness
An experimental C++ codebase for studying regret minimization with nonstandard estimators.
LLM exploitability audit
A Python toolkit for exact worst-case audits of language-model policies.
Safe observation capacity
Research code for active de-censoring in imperfect-information games.