Mar 2022 – Present · 4 yrs

• Market Microstructure Simulation: Engineered a novel data synthesis framework to generate high-fidelity tick data from standard 1-minute bars. This solved the critical data limitation problem for backtesting latency-sensitive strategies and enabled deep research into intra-bar market dynamics.
• Backtesting Infrastructure: Architected a high-fidelity, event-driven backtesting engine from scratch. The framework was designed to eliminate common statistical pitfalls by implementing Purged K-Fold Cross-Validation, embargoing, and dynamic, ATR-based slippage models.
• Machine Learning Integration: Engineered modules within the research framework to test and validate advanced models. Explored applications of time-series networks (LSTM) for predictive modeling and Reinforcement Learning for execution algorithm optimization.
• Strategy Research & Validation: Conducted systematic research across statistical arbitrage and momentum models. Utilized the backtesting framework to rigorously test hypotheses and, critically, to invalidate strategies suffering from subtle biases, ensuring a robust and truthful research process.
• Infrastructure Development: Designed the complete, end-to-end architecture for a systematic trading operation. This included robust data pipelines, a modular research environment, and components optimized for low-latency performance using Python and C++.