Purpose:
Set up an automated, scalable ML workflow using modern MLOps tools.
Technologies Used:
- Kubernetes (Minikube)
- Kubeflow Pipelines
- KServe
- AWS (EC2, S3)
- Python, Docker
Key Tasks:
- Created an ML pipeline covering data prep, training, and serving.
- Linked AWS services for model storage and secure access.
- Deployed the model with KServe for scalable inference.
Skills Applied:
- MLOps and pipeline automation.
- Cloud and container management.
- Secure, scalable ML deployment.
Result:
A working ML deployment on Kubernetes and AWS, built for efficiency and repeatability.
Link: Deploying Machine Learning Models with Kubeflow and KServe: A Comprehensive Guide | by Shijun Ju | Jun, 2025 | Medium
Code: shj37/Minikube-Kubeflow-ML-KServe-AWS-Project
All credit to iQuant for the original project design and code.
