Production multi-agent trading platform with rigorous walk-forward validation. TSMOM momentum (1.097 Sharpe) + GEX regime filtering. Interactive CLI, autonomous trade lifecycle, daily scheduler. Alpaca integration. Built on Microsoft AutoGen. Research-driven approach with statistical validation. Educational - not financial advice.

End-to-End Python framework implementing bias-adjusted LLM agents for human-like decision-making in economic games (Kitadai et al., 2025). Features persona-conditioned agent populations using Econographics data, multi-provider API integration, and Wasserstein distance validation against empirical benchmarks.

A comprehensive implementation of CBAM-STN-TPS-YOLO architecture for agricultural object detection, featuring convolutional block attention modules (CBAM), spatial transformer networks (STN), and thin plate spline (TPS) transformations. Includes cross-dataset evaluation on PGP, GlobalWheat, and MelonFlower datasets with statistical validation.

Secure Neural Network Model Watermarking

Machine learning pipeline for predicting used car prices using Kaggle’s Playground S4E9 dataset. Includes data cleaning, EDA, feature engineering, statistical analysis, and advanced regression models (LightGBM, XGBoost, CatBoost). Modular notebooks, visuals, and documentation included.

Research-grade implementation of Bidirectional ALT for shortest path problems — achieving up to 8× speedups on structured graphs with full statistical validation.

Calculates the potential maximum daily loss for an investment portfolio using three different financial risk methods (Value-at-Risk). The system checks these methods are accurate using statistical backtesting, helping set safe risk limits and meet financial rules.