Client: Orcapod Capital

Role: Machine Learning Engineer | Quant Engineer

Project Objective:

Developed a real-time algorithmic trading system that autonomously executes trades based on predictive models.

Techniques:

• Long Short-Term Memory (LSTM) networks for time series forecasting
• Statistical analysis and signal processing for feature engineering
• Reinforcement learning for optimizing trading strategies
• Multithreaded architecture for high-frequency execution
• Scheduled automation using cron jobs
• Pattern recognition and anomaly detection

Key Accomplishments:

1. Designed and implemented a deep learning-based trading bot that autonomously executes profitable long and short positions.
2. Developed a modular strategy engine powered by LSTM networks and advanced machine learning algorithms, enabling dynamic adaptation to market conditions.
3. Applied time series analysis and trend prediction techniques to train robust LSTM models on historical price data, resulting in a reliable and high-performing trading strategy.
4. Engineered cross-correlation features between technical indicators (MACD, RSI, PVT) to enhance the predictive power of the models.