Anyscale Courses

Taming Distributed AI Training with Ray and Datadog Observability

Discover how to make large-scale Ray training workloads observable and reliable using Datadog's proven techniques for monitoring, debugging, and optimizing distributed AI systems.

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

How Coinbase Uses Ray, vLLM and LiteLLM to Power Secure LLM Services

Discover how Coinbase built secure, production-grade LLM services using Ray, vLLM & LiteLLM for one of the world's most trusted crypto exchanges with distributed scaling & authentication.

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

How DataRobot Parallelizes Agentic Pipeline Searches with Ray and syftr

Discover how DataRobot's syftr framework uses Ray to optimize agentic LLM pipelines through massive distributed search, finding workflows 9× cheaper while maintaining accuracy.

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

AWS + vLLM - Building the Future of Open, Fast LLM Serving

Discover how AWS leverages vLLM for large-scale LLM inference, featuring deployment strategies, multi-accelerator clusters, and open-source contributions to optimize performance and costs.

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

High-Throughput Inference for Synthetic Data and Evals at Sutro

Discover how Sutro optimizes vLLM for massive-scale batch inference, covering custom implementation layers, performance profiling, and cost attribution for predictable synthetic data generation.

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

Efficient, High-Performance AI Inferencing with Intel Xeon 6

Discover how Intel Xeon 6 processors deliver cost-effective AI inference for enterprise applications, featuring AMX acceleration, enhanced memory bandwidth, and Ray integration.

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

End-to-End GenAI Orchestration with KubeRay

Discover how KubeRay unifies GenAI model lifecycles—from training to deployment—eliminating complex scripts and manual setup for streamlined ML workflows on Kubernetes.

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

SQL or Python? Why It's the Wrong Question in the AI Era

Explore how data and model infrastructure are converging in the GenAI era, examining the blending boundaries between SQL engines and Python frameworks for unified AI-driven systems.

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

Ray Agent Engine - Deploying AI Agents with Ray Serve

Discover how Apple engineers use Ray Serve as a framework-agnostic foundation for deploying scalable AI agents with multi-step reasoning, dynamic execution, and production resilience.

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

Getting Started with Distributed Training at Scale

Master distributed training techniques using PyTorch and Ray to scale ML models from single-GPU to massive clusters with DDP, ZeRO, and FSDP methods.

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

Scaling LLM Fine-Tuning with FSDP, DeepSpeed, and Ray

Master distributed LLM fine-tuning across GPU clusters using FSDP, DeepSpeed, and Ray with hands-on orchestration and memory management strategies for frontier-scale models.

• YouTube
• 1 hour 28 minutes
• Self-Paced
• Free Video

Why Ray Became a Distributed Computing Engine for Modern AI

Discover how Ray addresses the shift from I/O-bound to compute-bound AI systems, providing distributed execution for scalable, fault-tolerant Python workloads across CPUs and GPUs.

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

AI Open Source Stack Panel with vLLM, PyTorch, and Kubernetes

Explore open-source AI infrastructure foundations with industry leaders discussing vLLM, PyTorch, and Kubernetes collaboration, standards, and community-driven innovation.

• YouTube
• 1 hour 32 minutes
• Self-Paced
• Free Video

Bringing AI to the Physical World with Physical Intelligence

Explore how Physical Intelligence develops general-purpose AI models to enable any robot to perform any task, tackling hardware, software, and movement challenges.

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

Where AI Builders Shape What's Next - Ray Summit 2025 Keynote Day 1

Explore Ray's evolution, unified AI compute platforms, and major product updates including PyTorch Foundation integration, Azure partnership, and robotics advancements from industry leaders.

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