Course Wiki

Kubernetes

Orchestrating containerized model services: pods, services, scaling, and TensorFlow Serving.

The final ML Zoomcamp module serves models with Kubernetes. A trained TensorFlow model runs behind TensorFlow Serving; Kubernetes wraps it in a deployment, exposes it through a service, and distributes traffic across replicas. The course teaches the core object model — pods, deployments, services, ingress — at the depth needed to run model serving, not to administer clusters.

Scaling and load distribution are the operational payoff: when prediction traffic grows, Kubernetes adds replicas behind the same endpoint. The module completes the course arc — from a notebook model in module 2 to a horizontally scaled prediction service.

Taught in

• Machine Learning Zoomcamp

  • Module 10: Kubernetes and TensorFlow Serving
    • Overview
    • Introduction to Kubernetes
    • Deploying a simple service to Kubernetes
    • Deploying TensorFlow models to Kubernetes

• AI Dev Tools Zoomcamp

  • Module 4: DevOps and Observability for AI-Built Apps
    • DevOps and Observability for AI-Built Apps

• MLOps Zoomcamp

  • Module 07: Course Project
    • Course Project ## Related concepts

• Serverless Deployment

• Docker

• FastAPI

• Model Deployment