Anyscale Courses

Evolving Ray Core - Scalability, Reliability, and Compiled Graphs

Discover the latest advancements in Ray Core's architecture, including streaming generators, weekly releases, and the new Compiled Graphs backend for enhanced performance in accelerator-driven computing environments.

• YouTube
• 37 minutes
• Self-Paced
• Free Video

Scaling Ray to 10,000 NPUs - Huawei's Hyperscale Journey

Dive into Huawei's groundbreaking journey of integrating 10,000 Ascend NPUs with Ray, exploring technical innovations, observability solutions, and strategies for hyperscale AI infrastructure deployment.

• YouTube
• 20 minutes
• Self-Paced
• Free Video

Scaling AI at Autodesk with Ray and Metaflow

Discover how Autodesk combines Ray's distributed computing with Metaflow's workflow orchestration to process terabytes of data and train advanced 3D generative AI models at scale.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

From Data Engineering to MLOps: Building a Scalable ML Platform - Reverb's Evolution

Discover how Reverb transformed their data engineering team into MLOps, exploring scalable ML platform development, developer experience optimization, and effective monitoring strategies using Ray and Anyscale.

• YouTube
• 12 minutes
• Self-Paced
• Free Video

Scaling LLMs on Google Cloud - Synergy Between Ray, TPU, and GKE

Discover how to optimize LLM deployments using Ray, TPUs, and GKE on Google Cloud, focusing on performance optimization, memory management, and large-scale orchestration techniques.

• YouTube
• 17 minutes
• Self-Paced
• Free Video

Multi-tenant Data Processing with Ray - Building Industrial AI Platforms

Discover how Phaidra leverages Ray to build a multi-tenant data processing platform for industrial AI, focusing on reinforcement learning, efficient resource utilization, and secure customer onboarding.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

Fighting Fire with Algorithms - Lockheed's Reinforcement Learning-Based Wildfire Solution

Discover how Lockheed Martin uses deep reinforcement learning to revolutionize wildfire management through an AI-powered decision support system that achieves 80% containment success in simulated scenarios.

• YouTube
• 13 minutes
• Self-Paced
• Free Video

High-Performance AI Model Serving with Ray Serve - A Rubrik Case Study

Discover how Rubrik achieved high-performance AI inference using Ray Serve, exploring scalability solutions, fault tolerance strategies, and deployment techniques for handling millions of daily model evaluations.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

Pricing and Packaging AI Products for Scale

Master effective strategies for pricing AI products with insights on value-based models, market adaptation, and competitive positioning in the rapidly evolving artificial intelligence landscape.

• YouTube
• 11 minutes
• Self-Paced
• Free Video

The LLM-Cloud Synergy: How In-house LLM Teams Transform Cloud Development

Discover how NebiusAI leverages its unique position as both LLM developer and cloud provider to create more intuitive and efficient cloud services, with insights into the future of AI infrastructure.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

KubeSecRay - Fortifying Multi-Tenant Ray Clusters on Kubernetes

Discover how to secure multi-tenant Ray clusters using KubeSecRay, implementing robust Kubernetes security features for isolation, access control, and resource management in distributed computing environments.

• YouTube
• 16 minutes
• Self-Paced
• Free Video

Supercharging ETL and Analytics with Ray and Daft

Discover how Daft library enhances Ray clusters with distributed ETL capabilities, enabling seamless data processing, analytics, and ML/AI integration for scalable, high-performance workflows.

• YouTube
• 11 minutes
• Self-Paced
• Free Video

Distributed Mesh Processing and Geometric Deep Learning for Wind Prediction

Discover how Ray's distributed computing powers real-time wind prediction in urban design through mesh processing, geometric deep learning, and CFD data analysis for architectural innovation.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

From Spark to Ray - Data Processing Revolution with Daft

Discover how CSS revolutionizes machine learning infrastructure using Daft and Ray, showcasing real-world applications that streamline data processing and ETL workflows for enhanced efficiency and scalability.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

How Datadog is Transforming Time Series Forecasting with Toto

Discover how Toto, a Time Series-Optimized Transformer, revolutionizes forecasting through advanced architecture and trillion-point training data, enabling precise zero-shot predictions for observability metrics.

• YouTube
• 13 minutes
• Self-Paced
• Free Video