Production-grade causal uplift modeling on 14M rows, benchmarks S-Learner, T-Learner, and FT-Transformer challengers on the Criteo dataset, with Optuna tuning, MLflow tracking, FastAPI + Docker + Google Cloud Run serving, and a Streamlit dashboard.

Champion/challenger pricing framework — shadow mode, SHA-256 routing, SQLite quote log, bootstrap LR test, ENBP audit (146 tests)

AWS-oriented ML platform for fraud detection, churn prediction, anomaly detection, experimentation, monitoring, and risk inference.

Automated ML retraining with champion/challenger promotion, fairness regression gate, and one-command rollback. Built on MLflow Model Registry Aliases.

Scheduled ML retraining pipeline — DVC data versioning, MLflow experiment tracking, champion/challenger gating, and full audit trail. Auto-retrains weekly via GitHub Actions. Project 2/10 of MLOps portfolio.