Ray x DeepSpeed Meetup: AI at Scale
Anyscale and Snowflake are excited to host an evening focused on Ray and DeepSpeed, two of the most powerful open-source projects helping deliver large scale AI in production.
Join us for networking, lightning talks from AI experts from Snowflake, Anyscale and LinkedIn on how today’s most ambitious AI teams are scaling smarter and faster with Ray’s distributed computing framework for Python and DeepSpeed’s easy-to-use deep learning optimizations for large-scale model training.
5:30-6:00 Welcome & Networking
6:00-7:00 Lightning Tech talks:
DeepSpeed overview
SuperOffload: Novel offloading technique for Superchips
Arctic Long Sequence Training: Scalable and resource-efficient long sequence training
Muon Optimizer Support: Extend DeepSpeed (ZeRO) optimizer to enable Muon
DeepCompile: Compiler-driven optimization of distributed training
Snowflake ML: Using Ray for partitioning ML training
7:00-7:30 Q&A & Networking
Featured Speakers
Masahiro Tanaka, Software Engineer, Anyscale
Tunji Ruwase, DeepSpeed Lead, Principal Software Engineer, Snowflake AI Research
Zhipeng (Jason) Wang, Senior Manager, LinkedIn
Vinay Sridhar, Senior Product Manager, Snowflake
About Anyscale
Anyscale, the company behind Ray open source, is a fully-managed, enterprise-ready unified AI platform. With Anyscale, companies can build, deploy, and manage all their AI use cases, bringing transformational AI products to market faster. On the Anyscale platform, Ray and DeepSpeed work together seamlessly, enabling developers to scale distributed training with ease.
Join the Community!
DeepSpeed LinkedIn & Twitter / X
Join the Ray Community
Snowflake Developers on LinkedIn