Python Projects

Explore the current python projects I’m working on, ranging from ML/AI technologies to autonomous trading systems. Click on each project to learn more

Three Engines:
Automated Trading System

“Three Engines is a self-adjusting, Python-based microservice trading system that leverages momentum and volatility strategies, portfolio risk management, and API-driven execution for real-time deployment. It integrates yFinance, MongoDB, and the Alpaca Trading API to optimize market interactions.





Learn More

MarketPulse: Realtime Premarket Dashboard
(Work in Progress)

MarketPulse is a real-time stock market scraper and dashboard that tracks pre-market and after-hours trading for NASDAQ 100 and S&P 500 companies. It leverages Python-based scrapers, multithreading, and Dash to deliver an interactive, auto-updating view of stock movements and trends. Designed for traders and financial enthusiasts, MarketPulse ensures seamless monitoring with intelligent market-hour detection and robust handling of missing data.

Learn More

Argos-Hiragana Coding Environment

Argos Hiragana Coder is an NLP based utility that translates Python-like pseudo-code written in Hiragana into executable Python code using English syntax, leveraging the Argos Translate library. It helps Japanese learners to learn Python by automating the translation of keywords, comments, and function names, bridging language barriers in coding education.

Learn More

Upcoming Coding Project

This slot is reserved for an exciting new project currently in the works. More details will be updated soon!








Learn More

My WorkS in FinanCIAL Engineering

Key Projects & Research

A showcase of past machine learning and AI-driven finance projects, highlighting innovative solutions in market classification, algorithmic trading, and predictive modeling. While these projects are no longer actively maintained, they demonstrate key advancements in data-driven financial technologies

Machine Learning Market Classification Engine

• Developed an ML classification model for Russell 3000 industry sectors.
• Utilized Doc2Vec, NLTK, LSTM, GPT-4, Autoencoder, PCA, K-Means, GMM, UMAP.
• Achieved a 32% R² improvement over GICS benchmarks in low-market cap scenarios.

Carbon Allowance Price Prediction Model

• Built Python ML models for predicting EU & Chinese carbon allowance returns
• Used Empirical Mode Decomposition (EMD), BS-LSTM, Random Forest models.
• Achieved a Sharpe ratio of 4, outperforming baseline models.