A quantitative trading strategy backtester with an interactive dashboard. Enables users to implement, test, and visualise trading strategies using historical market data, featuring customisable parameters and key performance metrics. Developed with Python and Polars.

Production multi-agent trading platform with rigorous walk-forward validation. TSMOM momentum (1.097 Sharpe) + GEX regime filtering. Interactive CLI, autonomous trade lifecycle, daily scheduler. Alpaca integration. Built on Microsoft AutoGen. Research-driven approach with statistical validation. Educational - not financial advice.

Kiploks Trading Robustness Engine is an open-source TypeScript engine for deterministic backtest and walk-forward analysis (WFA) of algorithmic trading strategies, published as @kiploks/engine-* packages under Apache 2.0.

End-to-end ML system for prediction market trading — 521K markets, 78 features, 7 model architectures, walk-forward validation, live VPS A/B across 7 configs. Honest research-stop on alpha decay (NO-GO verdict). AFML methodology: Purged K-Fold, Deflated Sharpe Ratio, meta-labeling, focal loss.

Rigorous ML framework for alpha generation research. Demonstrates that apparent alpha from public market data is primarily overfitting. Key finding: IC flips from +0.036 to -0.084 when regularization is increased 100x.

Temporal partitioning and backtesting for time-correlated datasets

Motor de decisión ML para trading cuantitativo con validación walk-forward anti-leakage, triple-barrier labeling, XGBoost + Optuna, risk management para cuentas micro, human-in-the-loop y paper trading. Sistema completo Python 3.11+ de producción para NAS100 M5 con gating de modelos, costos realistas y kill switch.

Multi-model time-series forecasting with Bayesian Optimisation (Optuna TPE): SARIMA, Random Forest, XGBoost, LightGBM, Prophet, LSTM, and QuantileML probabilistic forecasts behind a unified ModelSpec protocol. Walk-forward validated; supports monthly, weekly, daily, and hourly data.

Kaggle Predict Future Sales — Public LB 0.87399 / Private 0.88034. LightGBM with 47 hand-engineered features. Documents 9 attempts including 7 rejected experiments.

Built a genuine next-day gold price forecasting system addressing the most common flaw in financial ML: same-day data leakage.

Multi-pillar L/S equity research on DoorDash (DASH): alt-data signals → GOV surprise forecast → revenue / contribution-margin / EBITDA chain → CAR event study. Pre-registered Q1 2026 prediction.

Systematic long-short equity & FIBRA pipeline for Mexican markets. Walk-forward LightGBM signal, SHAP attribution, Banxico regime conditioning. 172 tests · PDF tearsheet · Research report.

ML investigation into whether public data can generate profitable betting signals on European football markets

A leakage-controlled research framework for evaluating structured and semantic news signals in time-series asset return prediction using walk-forward validation.

ML pipeline that finds a real edge in binary options trading. LightGBM model trained on EURUSD/GBPUSD/USDJPY M5 data achieves 58.47% accuracy (breakeven: 55.56%) on 18 months of unseen data. Walk-forward validated. 15-min expiry strategy with confidence filtering.

Quantum Geometry and Topological Regime Detection in Financial Markets

IC-Gated Deployment Framework (ICGDF): two-stage ML deployment filter for cross-sectional equity prediction. Under review at QFE (AIMS Press).

📊Machine-learning decision system for Indian IPOs: probability model + backtested allocation strategy + FastAPI service + Streamlit dashboard.

Production-grade ML signal intelligence engine for quantitative trading. Powers real-time XGBoost inference across 100 S&P 500 tickers, 4-agent decision governance, algorithmic drift detection with automatic exposure scaling, and geopolitical risk overlay via live news APIs.

Academic ML research prototype for daily stock spike candidate screening with SEC-risk filtering, walk-forward validation, and recency-weighted candidate ranking. Not financial advice.