Uploaded October 2026 | Updated November 2026, 3 hours ago
As demand for high-performance AI clusters outpaces deployment speed, securing the underlying infrastructure has become a critical operational priority. Because traditional security boundaries blur at the data center fabric, the network must serve as the primary line of defense against modern threats. This session introduces a holistic blueprint for securing AI network infrastructure from the ground up. Walking through the complete network lifecycle, we will break down essential, multi-vendor security measures across Day 0, Day 1, and Day 2 phases. Attendees will leave with a practical framework to isolate backend GPU fabrics, mitigate data exfiltration risks, and protect massive AI workloads without sacrificing performance.
Saju Salahudeen:
Saju Salahudeen is a Principal IP Architect for AI & Cloud at Nokia and serves on the NANOG Workshop Committee. With over two decades in IP networking, specializing in multi-vendor network automation, gRPC services, and cloud architecture—he spends his time designing networks that scale with modern workload demands. A frequent workshop leader, Saju is always up to chat about network automation, gRPC workflows, open networking, or cloud infrastructure.
Reda Laichi:
Reda Laichi is a Principal Consulting Engineer in Nokia's IP Consulting Engineering team. Over more than 20 years he has followed the arc of network automation, from early SDN controllers to today's event-driven operations, delivering IP and data center networks for operators worldwide. His current work focuses on data center fabric design and automation, including the fabrics built for AI workloads.
As demand for high-performance AI clusters outpaces deployment speed, securing the underlying infrastructure has become a critical operational priority. Because traditional security boundaries blur at the data center fabric, the network must serve as the primary line of defense against modern threats. This session introduces a holistic blueprint for securing AI network infrastructure from the ground up. Walking through the complete network lifecycle, we will break down essential, multi-vendor security measures across Day 0, Day 1, and Day 2 phases. Attendees will leave with a practical framework to isolate backend GPU fabrics, mitigate data exfiltration risks, and protect massive AI workloads without sacrificing performance.
Saju Salahudeen:
Saju Salahudeen is a Principal IP Architect for AI & Cloud at Nokia and serves on the NANOG Workshop Committee. With over two decades in IP networking, specializing in multi-vendor network automation, gRPC services, and cloud architecture—he spends his time designing networks that scale with modern workload demands. A frequent workshop leader, Saju is always up to chat about network automation, gRPC workflows, open networking, or cloud infrastructure.
Reda Laichi:
Reda Laichi is a Principal Consulting Engineer in Nokia's IP Consulting Engineering team. Over more than 20 years he has followed the arc of network automation, from early SDN controllers to today's event-driven operations, delivering IP and data center networks for operators worldwide. His current work focuses on data center fabric design and automation, including the fabrics built for AI workloads.










