AI networking has become one of the industry’s most critical bottlenecks, limiting the speed, scalability, and efficiency of modern AI training and inference workloads. As clusters grow, operators require deterministic latency, extreme bandwidth, and open architectures that foster flexibility, interoperability, and rapid innovation. Upscale AI will explore how SONiC, an open-source network operating system, enables scalable and efficient AI networking. Upscale AI will also discuss practical strategies for adopting SONiC across the AI stack, and how cohesive, low-latency system design optimizes performance, accelerates deployment, and supports next-generation AI workloads.
Deepti Chandra is a product and strategy executive focused on open networking and cloud-
scale AI infrastructure. As VP, Product Management, Strategy & Marketing at Upscale AI, she
owns the product roadmap for high-performance, SONiC-based AI networking and drives
adoption across hyperscale and enterprise environments through ecosystem alignment and
partner execution.
A published author and patent holder in networking technologies, Deepti has spent her career
advancing the industry’s shift to open, cloud-scale platforms. At Cisco, she led hyperscaler-
focused open networking and Cloud PLM programs, including SONiC initiatives for large-scale
cloud environments. Previously at Hewlett Packard Enterprise, she developed growth strategies
for AI-as-a-Service and NeoCloud segments. She began her career in engineering and
technical product management at Juniper Networks.
Varun Vij is a software and networking architect focused on open networking technologies and
AI infrastructure platforms. Currently at Upscale AI as Director, Product Management, Varun
works across open network operating systems like SONiC, and next generation accelerator-
interconnect standards such as UALink, helping customers deploy scalable, high-performance
AI environments. Previously, he spent nearly eight years at Apple as a Software Architect
contributing to network virtualization, data plane development, and private cloud infrastructure.
Earlier, he held senior engineering roles at Juniper Networks supporting hyperscale and
enterprise data center platforms. With deep experience across software engineering and large-
scale networking operations, he brings practical insight into deploying SONiC for modern AI
networks. Outside work, he enjoys staying active, watching sports, and supporting his
daughter’s basketball team.