A Quick Overview of the Ray Libraries Built on Ray Core | Ray Summit Expo

At Ray Summit 2025, we took a walk around the expo floor to get a quick, engineer-led overview of the Ray libraries built on Ray Core. Ray is an open source AI Compute Engine for scaling AI and Python applications like machine learning. With Ray, developers can build and run distributed applications - without any prior distributed systems expertise. 🔗 Learn more about Ray: https://www.ray.io In this video, we stop by the booths to hear directly from the engineers building Ray and walk through how the Ray ecosystem fits together — from data processing and distributed training to model serving and Kubernetes. Whether you’re new to Ray or already using it in production, this walkthrough gives a quick look at what each Ray library is designed for and how teams use them in practice. Ray Libraries featured in this video: Ray Data - Scalable data processing for ML and AI workloads Ray Train - Distributed training and fine-tuning Ray Serve - Scalable model serving for online inference RLlib - Scalable reinforcement learning KubeRay - Running Ray on Kubernetes Chapters: 00:00 Introduction and Welcome 00:17 Ray Data 01:48 Ray Train 02:57 Ray Serve 05:18 KubeRay 06:00 RlLib