DEV Community

Avinash Mani Tripathi
Avinash Mani Tripathi

Posted on

Built a complete MLOps pipeline – looking for your feedback

I just finished building an MLOps project that I'd like to share with the community. The goal was to create a production‑ready pipeline that covers the entire ML lifecycle.

GitHub: https://github.com/avinashmnth2507-dev/mlops-house-price-predictor.git

Features:

  • ✅ Data versioning (DVC)
  • ✅ Experiment tracking (MLflow)
  • ✅ Hyperparameter tuning (Optuna)
  • ✅ Model serving (FastAPI)
  • ✅ Containerization (Docker)
  • ✅ CI/CD (GitHub Actions → GHCR)
  • ✅ Kubernetes deployment (Minikube)
  • ✅ Drift monitoring (PSI)
  • ✅ Monitoring stack (Prometheus + Grafana)
  • ✅ FinOps cost tracking

Known issue: The Grafana dashboard is currently showing "No data" due to a Prometheus scrape issue in my local Minikube cluster. I'm actively debugging – suggestions welcome.

I'm especially interested in feedback on:

  • The GitHub Actions workflow
  • Kubernetes manifests (deployment.yaml, service.yaml)
  • The drift monitoring implementation

Open to all constructive criticism. Thanks for looking!

Top comments (0)