Network Solutions Architect, AI Factory Services

Lenovo • Morrisville, NC

3d • Hybrid

About The Position

Lenovo seeks a highly experienced Network Solutions Architect to join the Hybrid Cloud Solutions and AI Offering Engineering team within SSG. This senior-level role designs, deploys, validates, and optimizes high-performance network infrastructure for Lenovo's AI Factory and GigaFactory service offerings, supporting both enterprise-scale and hyperscale GPU environments powered by NVIDIA Spectrum-X Ethernet and InfiniBand fabrics.

The architect will develop network reference architectures, deployment runbooks, performance validation procedures, and field-ready engineering documentation consumed by Lenovo Professional Services and Managed Services teams globally. This includes GPU cluster fabric design, BlueField DPU architectures, multi-tenant network isolation, high-performance RDMA and RoCEv2 deployments, and production-scale AI infrastructure supporting large distributed training and inference workloads.

The ideal candidate brings deep expertise in NVIDIA Spectrum-X, Quantum InfiniBand, and BlueField DPU technologies, along with hands-on experience architecting, deploying, validating, and troubleshooting large-scale GPU clusters. Experience supporting hyperscale AI infrastructure, high-density liquid-cooled environments, GPU cluster bring-up, infrastructure validation, performance tuning, and customer-facing technical engagements is highly desired.

Requirements

• Bachelor's or Master's degree in Computer Science, Electrical Engineering, Information Technology, or related discipline.
• 5+ years of experience designing and deploying high-performance networking solutions for AI infrastructure, HPC environments, GPU clusters, or Hyperscale cloud environments.
• Experience designing and validating large-scale network fabrics supporting AI and distributed compute workloads.
• Experience with customer-facing technical consulting, architecture reviews, or deployment leadership.

Nice To Haves

• Expert-level knowledge of NVIDIA Spectrum-X, Spectrum-4 SN5600 Ethernet, Quantum InfiniBand (HDR/NDR/XDR), BlueField DPUs, DOCA SDK, RDMA and RoCEv2.
• Experience supporting large-scale GPU cluster deployments and production AI environments.
• Deep understanding of GPUDirect, NCCL, AI workload communication patterns, and GPU cluster validation and performance tuning.
• Expertise in Adaptive Routing, SHARP, Congestion control, and Rail-optimized InfiniBand architectures.
• Strong experience with BGP, EVPN, VXLAN, MPLS, OSPF, IS-IS, and Leaf-Spine architectures.
• Experience designing lossless Ethernet environments utilizing PFC, ECN, DCQCN, and QoS.
• Knowledge of high-performance storage networking and end-to-end infrastructure optimization.
• Experience with Python, Ansible, Terraform, and REST APIs.
• Experience implementing infrastructure observability, telemetry, monitoring, and troubleshooting solutions for large-scale networks.
• Familiarity with data center operations, performance analytics, and capacity planning.
• Preferred certifications include: NVIDIA Certified Networking Professional, NVIDIA InfiniBand Specialist, Cisco CCNP Data Center, Cisco CCIE Data Center, OCI Networking Specialist, AWS Solutions Architect Associate.

Responsibilities

• Design GPU cluster network architectures for NVIDIA AI Factory environments utilizing Spectrum-X Ethernet (Spectrum-4 SN5600 + BlueField-3 DPUs) for enterprise deployments and Quantum XDR InfiniBand and Spectrum Ethernet for rack-scale GigaFactory deployments.
• Design and validate large-scale AI fabrics supporting RDMA, RoCEv2, GPUDirect, NCCL collectives, and high-bandwidth GPU-to-GPU communications.
• Develop rail-optimized InfiniBand topologies and Spectrum-X Adaptive Routing, SHARP, congestion management, and performance optimization strategies for AI training and inference environments.
• Contribute to architecture decisions supporting large-scale distributed AI and HPC workloads.
• Lead network bring-up, validation, and production-readiness activities for GPU infrastructure.
• Develop and validate RDMA/RoCEv2 configuration guides and deployment runbooks for AI and HPC environments.
• Establish performance baselines for RDMA throughput, RoCEv2 latency, GPU communication efficiency, storage and network throughput, and end-to-end infrastructure readiness.
• Design failure-domain isolation strategies and resilient network architectures for large-scale AI deployments.
• Support network architecture for high-density liquid-cooled GPU environments with power-aware design considerations.
• Diagnose and resolve complex InfiniBand, Ethernet, RDMA, and GPU workload performance issues.
• Analyze congestion, telemetry, traffic patterns, link utilization, routing behavior, and fabric health to identify bottlenecks and optimize performance.
• Support root cause analysis and remediation of network, storage, and infrastructure issues impacting AI workload performance.
• Develop operational runbooks and troubleshooting procedures consumed directly by Lenovo field teams and customers.
• Design multi-tenant networking architectures supporting AI Factory, NeoCloud, and managed-service provider environments.
• Implement namespace isolation, east-west traffic segmentation, secure tenant separation, and site resiliency aligned with NVIDIA Cloud Partner Reference Architectures.
• Design BlueField-3 DPU solutions utilizing DOCA for infrastructure offload, security services, observability, and service mesh capabilities.
• Collaborate with Lenovo engineering, product, ISG, and services organizations to validate designs against NVIDIA reference architectures and future hardware roadmaps.
• Work directly with customers, partners, and delivery teams to translate AI workload requirements into production-ready infrastructure solutions.
• Provide technical leadership during customer engagements, infrastructure deployments, escalations, and architecture reviews.
• Support global deployments through approximately 40-50% travel, including customer workshops, implementation support, solution validation, and executive-level technical discussions.

Benefits

• Hybrid Schedule on campus in Morrisville, NC. 3 days in office, 2 days work from home.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior