End-to-end MLOps platform for scoring sovereign economic stress risk across countries. Covers ingestion, validation, dbt warehousing, feature engineering, multi-model training (RF/XGBoost/LightGBM) with MLflow tracking, drift monitoring, a Streamlit dashboard, and a FastAPI serving layer containerized and Airflow-orchestrated.
python docker data-science machine-learning airflow time-series risk-analysis data-engineering data-analytics xgboost lightgbm mlops mlflow fastapi streamlit geoeconomics soverign-data
-
Updated
Jul 27, 2026 - Python