Sponsored Keynote: Inference and Sovereign AI: Scaling Cloud-Native... K. Angell & V. Caldeira (ASL)

This talk focuses on the deployment of AI inference pipelines at scale, addressing the challenges enterprises face in achieving high performance, resilience, and compliance. The speakers, Karena Angell and Vincent Caldeira from Red Hat, introduce Sovereign AI, a Kubernetes-native approach designed to orchestrate inference workloads effectively. They cover strategies for hardware-aware scheduling, dynamic scaling, multi-tenant resource management, and observability to support low-latency and high-throughput AI services across diverse infrastructure. Attendees will gain insights into containerizing AI models, optimizing the use of GPUs and accelerators, and ensuring compliance with regulatory requirements while running AI workloads in production.