​Food, Socializing, and Excellent talks from the PyTorch Ecosystem

​Venue: Engineering 2, Room 180, UCSC

• ​5:30 - 6:30 Food and Socializing

• ​6:30 - 7:00 Accelerating LLM Training and Inference with PyTorch on GPU

• ​7:00 - 7:30 PyTorch Devcontainers for AI/ML Workflow Optimization

• ​7:30 - 8:00 Talk 3

​Talk 1:
Accelerating LLM Training and Inference with PyTorch on GPU - We will be covering how to improve PyTorch training and inference performance by leveraging CUDA and tensor cores.
Presenter: Zeyuan (Faradawn) Yang, Techincial Marketing Engineer, NVIDIA

​

​Talk2:

​PyTorch Devcontainers for AI/ML Workflow Optimization - We will be looking at how you can use PyTorch Containers to avoid the usual frustrations of installing GPU drivers in your projects
Presenter: Steve Pousty, Principal AI Community Architect, Red Hat