This repository is an evolving MLOps sandbox focused on building reproducible, cloud-native ML pipelines.
This repository contains experiments focused on integrating Machine Learning workflows with cloud-native infrastructure and automation.
The objective is to build scalable, production-style ML pipelines rather than isolated notebooks.
- Build automated data preprocessing pipelines
- Implement model training workflows
- Integrate infrastructure provisioning via Terraform
- Automate deployment using CI/CD pipelines
- Add monitoring and observability
Data Source → Preprocessing → Model Training → Model Storage → API Deployment → Monitoring
- Data ingestion pipeline
- Training automation script
- Containerized ML model (Docker)
- CI/CD integration
- Cloud deployment (AWS / Azure / GCP)
- Logging & metrics
This repository reflects my interest in combining:
- Infrastructure as Code
- DevOps automation
- Cloud engineering
- AI/ML workflows
The focus is on building scalable and reproducible ML systems aligned with production engineering practices.
Build the image:
docker build -t ml-pipeline .