Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore the essential components and considerations for building self-managed AI networks in this conference presentation from the AI Engineer Summit 2025. Learn about new terminology specific to AI data centers, key power requirements for racks, and the specialized hardware needed including GPU servers. Discover the software stack requirements covering workload managers, GPU management, and programming tools, along with specific storage needs for AI workloads. Understand multi-tenancy options, uptime considerations, and the importance of telemetry and visibility in AI infrastructure. Dive deep into the critical differences between traditional data center networks and next-generation AI networking requirements. Gain insights from Paul Gilbert, a technology veteran with over 30 years of experience including 25+ years at Cisco Systems as Distinguished Systems Engineer and current tech lead at Arista Networks focusing on AI and data center solutions.
