Project List
– Data-Science Framework development
– Overall Market Overview
– Pre-market Stock Screener
– Stock Analysis Report
Data-Science Framework Development: Developed a full machine learning solution for the Kaggle Spaceship Titanic competition, applying the complete data science workflow—from data preparation to model deployment. The project involved extensive exploratory analysis to uncover key data patterns, automated preprocessing through pipeline construction, and iterative feature engineering to improve model accuracy. Classification models were built and optimized using Scikit-learn, with careful attention to data quality, variable transformation, and evaluation metrics. This project also produced a reusable framework for tackling future modeling problems, emphasizing reproducibility and structured problem-solving. <more info>
Overall Market Overview is a pre-market intelligence platform integrating multi-source ETL, LLM-driven news analysis, and interactive market visualizations to support data-driven trading decisions. Showcases skills in data engineering, BI, quantitative analysis, and applied AI. <more info>
Pre-market Stock Screnner: The Pre-market Stock Screener is a custom-built tool designed to automate and streamline the early-morning stock scanning process for day traders. Developed using Python with a Flask-based front end, it aggregates real-time data from multiple online sources to categorize and prioritize stocks based on actionable catalysts such as earnings reactions, news-driven moves, earnings surprises, second-day momentum plays, and overbought setups. By structuring raw market data into focused, strategy-aligned categories, the screener enhances decision-making speed and trade quality at the market open. The project tackled challenges in data collection, rate limiting, and signal filtering, while deepening the developer’s market domain knowledge and quantitative design skills. <more info>
Stock Analysis Report is an end-to-end stock analysis and trading decision support system built to streamline the identification of high-probability day trading setups. It generates detailed reports for individual stocks by combining structured market data with advanced natural language processing. Each report includes company context, earnings schedule, sector and market ETF performance, multi-timeframe technical analysis, quantitative behavior around catalysts, and real-time liquidity scoring. It also uses LLMs to analyze and summarize large volumes of financial news into highlighting key narratives. <more info>
