DLIT82183 | Thursday, March 19 | 8:00 a.m. - 9:45 a.m.

NVIDIA Run:ai is a comprehensive platform that brings advanced AI-native scheduling and resource management to GPU infrastructure, enabling enterprises to accelerate and scale AI operations efficiently, reduce costs, and accelerate innovation.

In this lab, you will learn the basics of NVIDIA Run:ai and how to create AI workloads. Designed for AI practitioners, this lab guides you through the practical steps of creating, launching, and monitoring interactive, inference, and training workloads using NVIDIA Run:ai.

Prerequisites

• Familiarity with the NVIDIA Run:ai platform
• Familiarity with containerised environments (e.g., Docker)
• General knowledge of AI/ML workflows
• Experience working with machine learning libraries and tools, such as PyTorch and Jupyter Lab

Details

Industry	All Industries
Technical Level	Technical - Intermediate
NVIDIA Technology	NVIDIA Run:ai
Intended Audience	Data Scientist

Key Takeaways

• Understand NVIDIA Run:ai workload types (Workspace, Training, Inference) and their appropriate use cases
• Submit and manage workloads through the Run:ai UI
• Set up and customise working environments for development, training, and inference
• Monitor workloads using dashboards, metrics, logs, and event history

Your Instructor

Juan Delgado - Senior Technical Instructor, NVIDIA

Juan is currently a Senior Technical Instructor for AI and Data Science at NVIDIA. Before joining NVIDIA, Juan worked at Coursera and Udacity developing AI/ML courses.