End-to-end quantitative trading system built on a 459,000-candle data pipeline with a 29-feature ensemble (XGBoost, RF, GBM, ET, Ridge). Validated via purged walk-forward CV with bootstrap confidence intervals and PBO measurement. Implements López de Prado's purged K-fold CV with triple-barrier labeling and embargo periods to eliminate label leakage in time-series ML models.
Full-stack AI app that restores color to grayscale images using the Zhang et al. deep learning model. A FastAPI backend runs a Caffe network — converting images through LAB color space, predicting 313 AB color bins, and returning a colorized JPEG in under a second.
Treepz • Toronto, ON

I'm a Computer Science student at York University with a deep interest in artificial intelligence and machine learning. I work with Python, scikit-learn, and pandas to build models that classify, predict, and cluster. I'm certified in ML and Generative AI, and I'm actively looking for opportunities to apply that knowledge to real problems at scale.