Quantitative Research | High-Frequency Trading | Machine Learning | Deep Learning
thibaut[at]uchicago.edu
I am a quantitative finance graduate student at the University of Chicago with a background in engineering, data science, and production machine learning. My work sits at the intersection of statistics, stochastic modeling, and machine learning for systematic trading.
- Statistical arbitrage and market‑making strategies in equities, crypto, and options
- Machine Learning and deep learning for signal discovery and validation
- Time‑series modeling
-
Cross-Stock Anomaly Research (Equities)
Research project on extending classical pairs trading to group-based equity strategies. The work uses multi-factor risk-model residualization, spectral embedding of return correlations, and machine learning-based clustering to define regularly rebalanced stock groups and study group-level mean reversion and momentum. -
Crypto Funding-Rate Carry Strategy
Delta-neutral funding-rate carry strategy on perpetual futures across multiple exchanges. The design incorporates funding-rate z-scores, open-interest-based liquidity filters, and macro risk regimes, with a custom vectorized backtester to evaluate robustness, drawdowns, and beta neutrality. -
Avellaneda–Stoikov Market Making Framework (Crypto)
Implementation of an inventory- and volatility-aware market making framework in Python, inspired by the Avellaneda–Stoikov model. The project analyzes spread control, inventory paths, and adverse selection on a crypto derivatives venue using large-scale trade-level backtests. -
Auction Outcome Prediction (Kaggle)
End-to-end pipeline for predicting ad auction outcomes with neural networks in PyTorch. Includes exploratory data analysis, missing-data handling (KNN imputation), feature engineering, and model tuning with F1-based evaluation on large-scale auction data.