Projects

Statistical analysis of 20,000 crash game rounds verifies the 97% RTP claim. But 179 rounds per hour means expected losses exceed 500% of wagers hourly.

Training RoBERTa to predict Hacker News success revealed temporal leakage inflating metrics. How temporal splits, calibration, and regularization fix it.

Analysis of 32,000 HN posts reveals 65% negative sentiment correlates with 27% higher engagement. Six-model comparison, power-law distributions, full data.

Build an open-source ML RSS reader with swipe interface. Uses MPNet embeddings and Thompson sampling for personalized feeds that escape the filter bubble.

Build a privacy-focused newsletter with Python, Cloudflare Workers KV, and Resend API. Zero tracking, zero cost, full control. Open source code included.

Cursor AI deployed YOURLS on Azure via SSH in 15 minutes: server config, MySQL, SSL, and a custom plugin. Full transcript and tutorial available.

Build the right mental model for gradients with this PyTorch visualization tool. 2D surface plots with gradient vectors show the direction of steepest ascent.

How I trained a YOLOv11 model to detect playing cards at 99.5% accuracy and built a real-time Monte Carlo blackjack odds calculator using computer vision.

A hands-on project predicting postprandial glucose curves with XGBoost, Gaussian curve fitting, and 27 engineered features from CGM data. Code on GitHub.

GPT-3.5 matched RavenPack's 41% returns in a sentiment analysis trading strategy using 2,072 news headlines. See the full backtest results and comparison.