Anyscale Courses

Enabling End-to-End LLMOps on Michelangelo with Ray

Explore Uber's extension of Michelangelo for end-to-end LLMOps, leveraging Ray for scalable development with hundreds of A100 GPUs and integrating open-source techniques for efficient custom LLM creation.

• YouTube
• 30 minutes
• On-Demand
• Free Video

Elevating DoorDash Model Training with Ray: From Good to Great

Explore DoorDash's journey in adopting Ray to enhance ML model training, addressing scalability, cost, and observability challenges while improving forecasting and training pipelines.

• YouTube
• 31 minutes
• On-Demand
• Free Video

OneDigital ML Platform: Unleashing the Power of Machine Learning at Scale

Explore OneDigital's journey in building a scalable ML platform using Ray and Anyscale. Learn about architecture, strategies, and retail use cases for advanced ML models.

• YouTube
• 27 minutes
• On-Demand
• Free Video

Hyperparameter Tuning with Ray Tune for Next-Gen Training Platform at LinkedIn

Explore LinkedIn's cloud-native deep learning platform and its use of Ray(Tune) for hyperparameter tuning in AI model training, aiming to democratize AI for engineers.

• YouTube
• 26 minutes
• On-Demand
• Free Video

The Ray Dataplane: History and the Cloudsort World Record

Explore the evolution of Ray's dataplane, from zero-copy object store to powering Ray Data, and learn how it broke the Cloudsort world record for cost-effective 100TB data sorting.

• YouTube
• 28 minutes
• On-Demand
• Free Video

Intellectual Property with GenAI: What LLM Developers Need to Know

Explore intellectual property issues in AI development, covering US IP laws, challenges in using and training models, and future legal trends affecting AI products and businesses.

• YouTube
• 34 minutes
• On-Demand
• Free Video

Anyscale Workspaces: A Scalable Interactive ML Development Environment with Zero Setup

Enhance ML development efficiency with Anyscale workspaces. Learn zero-setup development, debugging, scaling, and production deployment for LLM applications using the Anyscale platform.

• YouTube
• 28 minutes
• On-Demand
• Free Video

Deploying Ray Cluster on an Air-Gapped Kubernetes Cluster with Tight Security Control - Challenges and Solutions

Deploy Ray Cluster on air-gapped Kubernetes with tight security. Configure GCS for fault tolerance, set probes for high availability, overcome network policy challenges, and build encrypted TLS connections for compliance.

• YouTube
• 30 minutes
• On-Demand
• Free Video

Scaling Distributed XGBoost and Parallel Data Ingestion with Ray - FlightAware Case Study

Scale distributed XGBoost and parallel data ingestion for runway prediction using Ray and AWS. Learn to organize training data, configure fault-tolerant clusters, and optimize costs for processing terabytes of aircraft data.

• YouTube
• 30 minutes
• On-Demand
• Free Video

Large Scale Pathways Recommender Systems (PaRS) at Verizon

Explore Verizon's innovative multi-task learning recommender system, PaRS, leveraging Ray ecosystem to enhance customer experience through deep learning and efficient large-scale model training.

• YouTube
• 31 minutes
• On-Demand
• Free Video

Simplified Training, Tuning and Deployment of Foundational Models as a Service with Red Hat OpenShift

Simplify end-to-end lifecycle of foundation models using Red Hat OpenShift Data Science. Learn about cloud-native, scalable stack for training, tuning, and inferencing with integrated open-source components.

• YouTube
• 28 minutes
• On-Demand
• Free Video

How Snorkel Builds Interactive Enterprise ML Products Using Ray

Explore Snorkel's approach to building scalable, interactive ML systems using Ray for enterprise products. Learn about distributed data processing and in-memory techniques for optimal performance across diverse customer requirements.

• YouTube
• 33 minutes
• On-Demand
• Free Video

Ray for Large-Scale Time-Series Energy Forecasting to Plan a More Resilient Power Grid

Explore Ray's use in large-scale energy forecasting for power grid planning. Learn how Kevala leverages Ray Core, Ray Serve, and KubeRay to predict future grid conditions and assess risks.

• YouTube
• 28 minutes
• On-Demand
• Free Video

Heterogeneous Training Cluster with Ray at Netflix

Explore building a heterogeneous training cluster using Ray at Netflix. Learn to run distributed training jobs on mixed CPU/GPU instances and optimize cluster configuration for machine learning workloads.

• YouTube
• 31 minutes
• On-Demand
• Free Video

Building Samsara's Machine Learning Platform with Ray

Explore how Ray addresses challenges in building Samsara's ML platform, unifying cloud and edge AI development while improving data exploration and model training experiences.

• YouTube
• 30 minutes
• On-Demand
• Free Video