15-strategy algorithmic paper trading platform on AWS EC2 — systemd-supervised Python services, risk engine with kill-lines, market regime detection, and automated analytics pipeline

Quantitative strategy validation pipeline HMM regimes, walk forward cost aware backtesting

A Python framework for testing trading strategies against the ways backtests mislead: look-ahead audits, matched-exposure controls, and block-bootstrap significance tests. The tester is itself tested - a property fuzzer plus mutation testing (4 planted engine bugs, all caught). Includes three case studies of rejected ideas.

AI-powered multi-agent quant signal generation engine. Uses LangGraph to orchestrate 4 LLM agents (News Analyst, Trading Analyst, Risk Analyst, Manager) that collaborate to generate risk-adjusted BUY/SELL/HOLD signals using real-time news, vector memory, and backtesting.

Personal research project combining software development, behavioural analysis and quantitative review to transform discretionary trading decisions into an auditable dataset.

End-to-end automated crypto trading workflow featuring market scanning, signal generation, paper trading, risk management, Telegram alerts, PostgreSQL analytics, and Google Sheets reporting.

Cost-aware time-series momentum on a $20 IBKR account

Small-account systematic trading bot for Alpaca — built live, diagnosed a losing strategy with real backtests, and rebuilt it.

AI multi-agent system for stock market signal generation using LangGraph, GPT-4, and Qdrant vector search. Achieved 42.8% backtest return vs. 24.5% buy-and-hold, 78% win rate on high-consensus signals. 🥇 Best Use of AI/ML, UB Hacking 2024.

Advanced IDX Market Intelligence & Screener Platform featuring AI-powered Reasoning, Deep Broker Flow Detection, and Automated Trading Journal.

Automated multi-asset mispricing bot for Kalshi BTC/ETH price-level markets — log-normal pricing, adaptive vol calibration, Kelly risk sizing, full replay/audit trail.

Event-driven trading backend — NestJS microservices over TCP + Redis, BullMQ scheduling, Postgres persistence. Sentiment scoring → risk gating → execution, running on a simulated market. docker compose up to run.

Sanitized public case study of AlphaQuant V12: systematic trading architecture, risk governance, QMS testing, safe demo code, and CI.

Public docs for Coil (coil.trade) — an agent-native, long-only trading system: scanner + dashboard + engine, run in your own AI agent (built for Claude Code) against your own broker's MCP. Docs only — not an MCP server.

Cross-sectional momentum backtest (12-1, monthly rebalance) on the 9 original SPDR sector ETFs — hand-rolled engine, net-of-cost results, factor regression + deflated Sharpe + block bootstrap.

High-performance C++20 order book engine with REST API, React web terminal, LOBSTER replay, and online ML pipeline.

Quantitative AI hedge fund platform: Flask backend, ML/RL trading models, React web and React Native mobile clients.

C++ limit order book with price-time priority matching, thread-safe concurrent access (ASan-verified), and full latency percentile analysis (p50-p99.9). 12/12 correctness + 5/5 concurrency gates.

Crypto trading bot for Kraken: Optuna-tuned strategies, walk-forward backtesting, live Fly.io deployment. Concluded research project.