I will design and build an end-to-end Machine Learning pipeline for financial, trading or crypto time-series data.
The work may include data preparation, feature engineering, target design, time-aware train/validation/test splits, model training, evaluation, inference and result analysis.
I focus on correct methodology for time-series ML: no random data leakage, no future information in features and validation that reflects real-world model usage.
Depending on the project, I can deliver:
• Python ML pipeline
• Feature engineering
• CatBoost / XGBoost / LightGBM / sklearn models
• Time-series validation
• Metrics and diagnostics
• Inference pipeline
• Clean reusable code and documentation
Suitable for crypto, financial markets, forecasting, classification and analytical systems.
