264001
eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.eyJpc3MiOiJid2V2ZW50cy1wdWJsaWMtYXBpIiwiYXVkIjpbInB1YmxpYy13ZWIiXSwic3ViIjoiaHR0cHM6Ly93d3cuZ29vZ2xlY2xvdWRldmVudHMuY29tIiwiaWF0IjoxNzc1MDUzNTgxLCJuYmYiOjE3NzUwNTM1ODEsImV4cCI6MTc3NTA1NzE4MSwiY29udGV4dCI6eyJwYXlsb2FkIjp7ImlkIjoiIiwicmVnVHlwZU5hbWUiOiIiLCJldmVudF9pZCI6IjI2NDAwMSIsImNfMzI5MDQxNiI6IiJ9fX0.92qh8Mure7pW2oh2qCe_H4DUBHxXn1RTs2gUkyAFvC0
phase 5 session info modal
Day 3 – April 24, 2026
Breakouts
BRK1-067 • Compute, Migration • Introductory
CPU infrastructure for the age of inference: Lessons from industry leaders
location_on Mandalay Bay F
schedule 8:30 AM - 9:15 AM
General-purpose compute is emerging as the critical engine for the age of inference. Join this session to learn firsthand from industry leaders who have successfully implemented a workload-optimized approach to power their AI stacks. We’ll cover topics ranging from high-efficiency CPU-based inference to the massive data preprocessing required for modern generative workloads. Learn how to build an AI-ready infrastructure that eliminates bottlenecks and minimizes latency across your entire pipeline. We’ll break down the emerging architectural patterns that provide the flexibility to innovate at scale, so your infrastructure can scale seamlessly without being locked into a single path. Read more
This agenda widget contains logic for handling same sessions available at multiple times
GTM Tracking Header name present
Session
contains logic for gtm tracking when modal open