- Python (Deep Learning, Data Preprocessing)
- TensorFlow / Keras – Model building
- MLflow – Experiment tracking, model logging, production-grade tracking
- DVC (Data Version Control) – Lightweight pipeline orchestration & experiment tracking
- Docker – Containerization for deployment
- GitHub Actions – CI/CD automation
- AWS (EC2, ECR) – Cloud deployment infrastructure
-
Configuration Management –
config.yaml,params.yaml -
Entity & Component Updates – Define data entities and ML components
-
Pipeline Orchestration – Update pipelines in
main.pyanddvc.yaml -
Experiment Tracking – MLflow logging with local/DagsHub tracking URI
-
Version Control – DVC for dataset, pipeline, and reproducibility
-
Deployment
- Dockerize the application
- Push image to AWS ECR
- Deploy on AWS EC2 via GitHub Actions
- End-to-end Kidney Disease Classification pipeline
- Experiment tracking (MLflow + DagsHub)
- Pipeline reproducibility (DVC)
- Cloud-native deployment using AWS, Docker, GitHub Actions