AI on GKE is a collection of examples, best-practices, and prebuilt solutions to help build, deploy, and scale AI Platforms on Google Kubernetes Engine
Manifest file to deploy all necessary K8s resources for it in a single command
Unit tests
E2E tests (i.e., jobsets with different specs for different scenarios you can deploy on a cluster to manually test e2e)
README with documentation for how to test, build, and deploy it
(I removed the SSL cert and private key I used for testing of course, so if you want to test yourself you'll need to replace the placeholders with your own)
This is an admission controller with a mutating Job webhook to inject nodeSelectors based on deployment specific environment variables.
The PR contains:
(I removed the SSL cert and private key I used for testing of course, so if you want to test yourself you'll need to replace the placeholders with your own)