Don’t miss out! Join us at the next Open Source Summit in Seoul, South Korea (November 4-5). Join us at the premier vendor-neutral open source conference, where developers and technologists come together to collaborate, share knowledge, and explore the latest innovations and advancements in open source technology. Learn more at https://events.linuxfoundation.org/
Keynote: Zero to 50 ExaFLOPS in Under a Year: Lessons from the Trenches – Hagay Lupesko, SVP, AI Inference, Cerebras Systems
The Cerebras Wafer Scale Engine is the world’s largest chip—50× bigger than a GPU and capable of over 100 petaFLOPS of AI compute. In less than a year, we scaled Cerebras Inference, powered by clusters of the Wafer Scale Engine, to more than 50 exaFLOPS—delivering the world’s fastest AI inference service for leaders like Meta, Mistral, and IBM. But the real story isn’t just speed—it’s scaling fast, and to a massive scale.
In this keynote, we’ll share lessons from that journey: what breaks (and when), why system design beats model tweaks, and how to build observability, security, and culture that can survive at exascale. Join us for the unvarnished, hard-earned insights on scaling AI inference—without having to learn them the hard way.
