Machine learning for data science, athletic performance, and quantitative finance.
GTM Systems & Analytics Intern @ FundRecs · BSc Science, University College Dublin (2026) · International 400m sprinter for Ireland (PB 46.95s).
- medalist — Harness for AI agents to autonomously solve tabular data-science competitions end-to-end; live agent placed top ~16% (est.) on a finished Kaggle Playground competition with zero human involvement.
- OptimalAthlete — ML system predicting 400m sprint times (R²=0.84), trained on 18 months of personal race and training data.
- GrowthHog — Systematic equity screener for 170+ global tickers: lifecycle-aware scoring, sector health, insider tracking, and portfolio optimisation, running weekly via cron on Polygon.io data.
- TradeMetrics — Pairs trading and portfolio analytics system (Sharpe 1.35), built on 12 months of Interactive Brokers data.
- HorseLay (private) — Automated lay-betting system for the Betfair exchange: price-shortening signal on flat racing, full backtesting engine, and live bot. Backtested on GB/IE/FR data across 2015–2024 (90.9% strike rate, small validated sample).
Python · R · SQL · scikit-learn · XGBoost · Streamlit