Open-source AI-driven quantitative trading platform for crypto, stocks, and forex with backtesting, live trading, market data, and multi-agent research.

Multi-symbol AI trading agent — Claude AI + FastAPI + Next.js + MetaTrader 5

ATOMIC MESH — Distributed deterministic HFT market-making engine. Avellaneda-Stoikov strategy with sub-microsecond C++ hot-path (575ns), event-sourced architecture, VPIN toxicity detection, QUIC mesh transport, real-time dashboard. Rust + C++17 FFI. Live on Binance.

Machine learning overlay for SPY using ^GSPC regime signals, Jump Model labels, and XGBoost. Focused on downside protection, recovery timing, and risk-adjusted improvement under a unified long-sample backtest.

BSc thesis: Genetic Algorithm for cardinality-constrained portfolio optimization on US equities (2000–2025). Compared against MVO and 1/N benchmarks using a rolling-window out-of-sample evaluation.

Quantitative risk & performance analysis of U.S. equities. Sharpe ratio, VaR, drawdown, momentum strategy backtest, and minimum variance portfolio optimization.

Python toolkit for backtesting trading policies, generating portfolio actions, and tuning strategies with walk-forward validation.

Designed a two-stage, end-to-end differentiable trading system in TensorFlow that generates and refines daily portfolio weights across ~800 assets using technical indicators. A simulated trading loop feeds back KPIs, enabling direct optimization of equity and Sharpe ratio via gradient-based training. It runs on Quantiacs.

Subscribe to TradeMAV ntfy notifications and forward parsed signals to a broker or webhook.

A Python-based financial CLI that tracks real-time portfolio diagnostics and mathematically stress-tests investments against historical Black Swan market crashes using Monte Carlo Jump-Diffusion models.

Scalable algorithmic trading backtesting engine for multi-asset strategies with automated reporting across NIFTY 50 equities.

crypto trader bot

🤖 AI Trading Bot 2026: MetaTrader 5 & GPT-4 API | Next.js Dashboard

Multi-task Deep Learning suite for Nifty 50 stock forecasting. Simultaneously predicts price returns and market direction using LSTM, GRU, and Transformers with bias-corrected weighted loss.