Quant-minded software builder working at the intersection of markets and code — from stochastic-volatility models to shipped, full-stack products. I like building things that are both rigorous and real.
🎯 Seeking a Summer 2027 Software Engineering internship · open to quant / fintech too 🧰 Python · TypeScript · React/Next.js · Postgres · a healthy obsession with validating my own work
📈 TradeGoons — full-stack AI trading app (shipped & live)
A paper- and real-trading journal with live market data, AI coaching, options chains, portfolio analytics, and a risk-visualizing UI. Installable as a phone app. Stack: Next.js · TypeScript · Prisma · Supabase (Postgres) · Vercel · Gemini API
🌊 soxl-vol-surface — options volatility research
IV rank & percentile signals, GARCH volatility forecasting, Black-Scholes Greeks, a 3D volatility surface, and a Heston stochastic-volatility model that reproduces the implied-vol smile a constant-vol model can't.
🔬 momentum-factor-backtest — strategies + honest validation
A cross-sectional momentum backtest, then the part most people skip: a Fama-French factor regression (which showed the strategy has no alpha — just momentum-factor exposure) and a cointegration pairs stat-arb study validated with out-of-sample splits, transaction costs, and a multiple-testing correction.
🛡️ leveraged-etf-risk-lab — portfolio risk analytics
Sharpe/Sortino, beta, drawdown, and stress tests for a leveraged-ETF book, plus a Hidden Markov Model that classifies the market into bull/bear regimes from return patterns alone.
🛰️ new-space-radar — event study
Measures cumulative abnormal returns (CAR) around launches and contract news for space-sector equities, with significance testing.
- Build strategies and try to break them. A pretty backtest is the start of due diligence, not the end — out-of-sample tests, cost modeling, and multiple-testing corrections are non-negotiable.
- Ship fast, then harden. I use modern AI tooling to move quickly, then dig into the code so I actually understand what I shipped.
- Make it legible. Every repo has a README that explains the why, the method, and the result.
Python · pandas · NumPy · statsmodels · SciPy · TypeScript · React · Next.js · Prisma · PostgreSQL · Supabase · Vercel · Git
- LinkedIn: https://www.linkedin.com/in/anggun-soracca-jaya-51a329236/?skipRedirect=true
- Email: [add your email]