Consulting Engineer – AI Networking & Fabric Architecture

The AI Infrastructure Consulting Engineer – AI Networking & Fabric Architecture is responsible for implementing, validating, and optimizing high-performance network fabrics that support accelerated computing and distributed AI workloads. This client-facing, billable role provides high-quality deliverables and ensures solutions are implemented as designed to the client’s satisfaction and approval. The engineer translates workload, topology, performance, and operational requirements into reliable network configurations for GPU clusters and adjacent data-center infrastructure. What You Will Do · Manage time and expense to meet or exceed expectations defined in the Statement of Work, proactively escalating potential scope, schedule, or budget issues as soon as they are known. · Implement and validate high-performance AI network fabrics using InfiniBand, Ethernet, RoCE, or related technologies. · Configure and troubleshoot lossless networking, congestion management, quality of service, routing, telemetry, cabling, optics, and fabric-management components. · Support GPU-to-GPU, node-to-node, storage, management, and client network connectivity required for distributed training and inference environments. · Execute installation, configuration, burn-in, performance validation, troubleshooting, documentation, and knowledge transfer in customer environments. · Provide technical input for SOWs, ROMs, LOEs, implementation plans, and solution validation when requested. · Create and maintain accurate project documentation, as-built configurations, test plans, work logs, and client handoff materials. What We Expect of You · Bachelor's degree and 3 year’s experience delivering complex network or data-center infrastructure solutions, including planning, implementation, integration, and support OR 7 year’s delivering complex network or data-center infrastructure solutions, including planning, implementation, integration, and support. · Hands-on experience implementing and troubleshooting enterprise data-center networks. · Working knowledge of Ethernet switching, routing, VLANs, LAG/MLAG, BGP, cabling, optics, telemetry, and network validation. · Understanding of lossless fabric concepts, including priority flow control, explicit congestion notification, congestion management, buffering, and quality of service. · Experience with InfiniBand, RoCE, RDMA, GPUDirect RDMA, or comparable high-performance networking technologies. · Ability to interpret network designs, port maps, bills of materials, cable matrices, and implementation plans. · Ability to build and execute test plans aligned to assigned project deliverables and the Statement of Work. · Strong written and verbal communication skills with the ability to effectively interact with leadership stakeholders and clients. · Proven ability to create accurate as-built documentation and transfer technical knowledge to clients. · Demonstrated attention to detail, analytical problem solving, and the ability to balance competing priorities while meeting deadlines. · Ability and willingness to work beyond regularly scheduled shifts, select weekends, or overtime as necessary to meet operational needs. · Ability and willingness to travel as needed to CDW locations or client sites. · Valid U.S. driver's license. · Experience with NVIDIA Spectrum-X, Spectrum Ethernet, Quantum InfiniBand, Cumulus Linux, SONiC, or comparable data-center networking platforms, is a plus. · Experience with NVIDIA UFM, NetQ, telemetry, automation, benchmarking, or fabric-management tooling, is a plus. · Familiarity with rail-optimized topologies, distributed training communication patterns, and AI storage-network requirements, is a plus. · Preferred NVIDIA certifications aligned to the role include NVIDIA-Certified Professional AI Networking (NCP-AIN) and NVIDIA-Certified Professional AI Infrastructure (NCP-AII), is a plus. Pay range: $77,250 - $108,265