I will perform a technical audit of your Machine Learning pipeline for trading, financial or crypto time-series data.
Many market ML models show strong backtest results because of hidden leakage, incorrect validation, target contamination or repeated optimization on the same historical period.
I will review your pipeline and identify methodological risks.
The audit may cover:
• Future-data leakage
• Target leakage
• Incorrect train/test splitting
• Random CV on time-series data
• Feature contamination
• Threshold overfitting
• Excessive feature mining
• Backtest overfitting
• Unrealistic inference assumptions
• Walk-forward validation design
• Metric selection and interpretation
You will receive a clear technical report explaining what is reliable, what is risky and what should be changed.
