EMR API – DevOps Practical Project

This is a small Electronic Medical Records (EMR) API I worked on to improve my DevOps skills. I wanted to take a simple Python API and make it production-ready using Docker, Kubernetes, and proper monitoring. The whole idea was to practice how real-world healthcare applications are packaged, deployed, and observed.

What I Built

I containerized the API using Docker, wrote the Kubernetes manifests myself, and added Prometheus so I can scrape metrics from the service. I set this up as part of my journey to understand how DevOps and MLOps pipelines are built end-to-end. This helped me see how an app moves from local development → container → cluster → monitoring.

Tech Stack I Used

Python – Basic API logic

Docker – Image building and running the service locally

Kubernetes (k8s) – Deployment and Service files for running it in a cluster

Prometheus – Collecting metrics from the API

Grafana – For dashboarding (optional for now)

How I Run the Project

When I’m testing locally, I normally use Docker Compose:

docker-compose up --build

For Kubernetes, I apply the files in the k8s folder:

kubectl apply -f k8s/ kubectl get pods kubectl get svc

Prometheus will pick up the metrics based on the config in prometheus.yml.

What I Learned

Working on this project helped me understand:

How to containerize a Python API cleanly

How Kubernetes Deployments and Services work together

How to expose and scrape metrics with Prometheus

How DevOps workflows support real applications like EMR systems