A Deribit options trading framework that finds mispriced crypto options by comparing market-implied probabilities against historical statistical forecasts. Automatically flags overvalued tail risk and skew to structure delta-neutral option spreads.

Building a cointegration pairs-trading strategy, then systematically dismantling every bias until only the honest result remains.

Volatility-aware ML system for statistical arbitrage with the purpose of learning when a pair spread is actually tradeable after costs, risk, and regime effects.

Production-grade framework for prediction market signal discovery using Bayesian inference, regime detection, and rigorous walk-forward backtesting. 122 unit tests, 85% code coverage.

Pairs trading framework that models spreads as Ornstein-Uhlenbeck processes and derives optimal entry thresholds by solving the Kolmogorov Backward Equation — replacing conventional fixed z-score heuristics with mathematically grounded, dynamically calibrated signals.

Statistical arbitrage research framework — equity pairs, crypto pairs, and funding rate carry strategy