As a distributed compute framework for AI applications,
Ray.io shares several similarities with Kubernetes, but takes a different approach to security. Ray has grown in popularity in recent years, and deploying it on Kubernetes with KubeRay is a popular choice that provides flexibility and configurable orchestration. In this talk, we’ll discuss recent security research that found misconfigured Ray clusters exposed to the Internet, and we’ll cover what options you have for hardening Ray deployments with Kubernetes. Can Kubernetes shore up security where Ray expects it to be configured by the user?
