Network Infrastructure Architect

SambaNova is a leader in next-generation AI infrastructure, delivering a full-stack inference platform for customers worldwide. At the core of SambaNova's technology is the RDU (Reconfigurable Dataflow Unit) — a chip built on a dataflow architecture rather than the traditional GPU model. Its decode performance is especially strong for agentic workloads like multi-turn agents, code generation, and long-running applications. RDUs are packaged into SambaRack, rack-scale hardware that lets customers deploy state-of-the-art models with better performance, greater energy efficiency, and faster time to value.

About the role

SambaNova is accelerating from rack-scale to global-scale AI compute clusters. We're looking for a Network Architect join us on this exciting new chapter and own this critical layer end to end.

Responsibilities

In this role, you'll be architecting the network fabrics that power hyperscale AI clusters, from RDU-accelerated servers to global-scale deployments, owning the fabric, interconnect, and spine-leaf design decisions that shape how our compute platforms are built. You'll partner closely with hardware engineering, supply chain, and customer-facing teams to translate workload requirements into platform roadmaps, taking a vendor-agnostic approach that avoids lock-in. Your architecture decisions will directly shape the products we ship and how fast the world's largest AI clusters come online.

Required Qualifications

• 12+ years designing or architecting infrastructure for hyperscale, AI, HPC, or large-scale data center environments
• Deep working knowledge of AI cluster networking, including InfiniBand, RoCEv2, Ethernet-based AI fabrics, 400G/800G interconnects, and GPU-to-GPU east-west traffic patterns
• Experience with lossless Ethernet designs, including QoS, ECN, PFC, congestion management, buffer tuning, and failure-domain isolation
• Strong understanding of spine-leaf architectures and associated technologies (VXLAN, EVPN, BGP, ECMP, underlay/overlay design, network automation)
• Familiarity with high-speed optics and cabling for AI clusters, including 400G/800G transceivers, DAC/AOC, fiber topology, and link budgets
• Experience translating customer and workload requirements into product requirements, platform roadmaps, and architecture tradeoffs
• Vendor-agnostic approach to platform selection and interoperability across switch, NIC, optical, and fabric ecosystems
• Ability to partner across hardware engineering, supply chain, operations, and customer-facing teams to bring platforms from concept through deployment

Preferred Qualifications