Quantitative researcher and trading-systems engineer. Vietnam equities and global FX / gold (XAUUSD). I build the whole pipeline: point-in-time data, signal research, walk-forward validation, execution, and the risk layer that stops it when the research turns out to be wrong.
The production system I run day to day is private. These repositories are self-contained components extracted from that work, rebuilt so they stand on their own with tests and documentation.
| Repository | What it is |
|---|---|
| backtest-engine | Backtesting with enforced execution lag, turnover-based costs, purged walk-forward CV, and trial-aware overfitting diagnostics (deflated Sharpe, PBO). |
| trading-risk-monitor | Real-time drawdown tracking and prop-firm rule enforcement. Equity-based, static and trailing floors, crash-safe session state. Live demo. |
| mt5-python-bridge | MetaTrader 5 connectivity: reconnection that detects a dead link, broker symbol resolution, filling-mode fallback, timezone-correct history sync. |
217 tests, ~90% branch coverage across the three. CI on Linux, Windows and macOS, Python 3.10-3.12.
A theme runs through all three: the interesting engineering is in refusing to produce a flattering number. The backtester raises rather than allow same-bar execution. The risk monitor measures on equity, not balance. The MT5 bridge surfaces partial fills instead of counting them as success.
Python · NumPy · pandas · FastAPI · pytest · MQL5 · MetaTrader 5 API
Open to remote quantitative research and trading-systems roles.
pnhongthai098@gmail.com