I will build python machine learning models and mlops pipelines

I developed an end-to-end ML & MLOps project to predict late deliveries using real Olist E-Commerce data.

PROJECT WORKFLOW & OUTPUTS

1. Data Integration | Python, Pandas
2. Joined related tables to create a unified dataset.
3. Target Engineering & Splitting
4. Created Late/On-Time labels and train/validation/test sets.
5. EDA & Feature Engineering
6. Analyzed delivery patterns and built preprocessing pipelines.
7. Model Training | Scikit-learn
8. Trained classification models and evaluated prediction metrics.
9. MLflow | Experiment Tracking
10. Recorded training runs, metrics, and model artifacts.
11. DVC | Version Control
12. Managed data and model versions.
13. Great Expectations | Validation
14. Defined data quality checks and input constraints.
15. FastAPI | Model Serving
16. Built /predict, /batch-predict, /health, /model-info, and /metrics endpoints.
17. Docker | Containerization
18. Configured Dockerfile and Docker Compose.
19. Pytest & GitHub Actions | CI/CD
20. Prepared automated tests and CI workflows.
21. Monitoring & Logging
22. Added request, error, latency, and prediction tracking.

DELIVERABLES

Python code, trained models, evaluation outputs, API, tests, and documentation.

Let's discuss your ML project!