I'm a senior double major in Mathematics and Computer Science at Sacred Heart University, pursuing a career in Quantitative Finance. In my spare time, I love to work out and cook!
I am currently converting a large USPIS FOIA file (200,000+ reported attacks) from a raw, non-analysis-friendly format into a clean, structured, analytics-ready dataset. After converting the file, I am analyzing attack patterns to produce actionable, location-specific insights that help leadership communicate impact, improve awareness, and support outreach to policymakers—especially in the hardest-hit congressional districts and states.
- Transform raw FOIA files into a usable dataset (consistent schema, validated fields, reproducible process).
- Quantify and localize attack trends across geography and time.
- Identify hardest-hit areas (district/state-level summaries and rankings).
- Create clear, shareable outputs (tables, figures, and brief-ready summaries) to support media coverage and congressional outreach.
SynTrade is a hybrid multi-agent trading system that combining sentiment analysis and technical indicators. It introduces a validation layer that verifies heterogeneous signals before execution, improving risk-adjusted performance metrics.
Stack:
- Python • Backtrader • LightGBM • Google Gemini API • pandas • NumPy • Matplotlib
- APIs: Finnhub • FRED • NewsAPI.ai • Alpha Vantage • yfinance
Highlights:
- Modular multi-agent trading system with a validation/critic layer before execution
- Backtrader backtesting + baseline comparisons (buy-and-hold, technical-only)
- Multi-source data pipeline (news + macro + fundamentals + technicals)
- LightGBM models + Gemini LLM integration for sentiment/credibility scoring
- Risk controls + logging/analytics (stops/exits, decision logs, performance visualizations)
This project compares Monte Carlo and Black–Scholes option pricing under identical geometric Brownian motion assumptions, using real market data to demonstrate the convergence.
Stack:
- Python • NumPy • pandas • SciPy (stats, optimize) • yfinance
Highlights:
- End-to-end Black–Scholes vs. Monte Carlo call option pricing (risk-neutral GBM)
- Live inputs via yfinance
- Implied volatility calibration from market prices
- Vectorized Monte Carlo with standard error + 95% CI reporting
