Job Description
Job Description
About the Company
\n
\n
GMI Cloud is a fast-growing, AI-native infrastructure company delivering high-performance GPU compute, inference services, and infrastructure for AI agents.
\n
\n
Following 8x ARR growth, GMI Cloud continues to scale rapidly across the U.S. and APAC. As a Reference Platform NVIDIA Cloud Partner (NCP) and a validated leading NCP across both markets, we power production AI for leading AI-native companies including Fireworks AI, Cartesia, Reflection, and OpenRouter.
\n
\n
From large-scale compute to optimized inference and agentic workloads, GMI Cloud gives AI teams the infrastructure they need to build, deploy, and scale on one unified cloud.
\n
\n
One cloud for compute, inference, and agents.
\n
\n
About the Role
\n
\n
We are seeking an experienced Network Architect to join the GMI Global Infrastructure team. You will be responsible for leading the design, deployment, and day-to-day operations of high performance network infrastructure supporting GPU-accelerated AI/ML workloads. This role requires deep, hands-on expertise with InfiniBand and RoCE, high-throughput low-latency fabrics, network troubleshooting, performance tuning, and cross-functional collaboration with various stakeholders.
\n
\n
Preferred Location: US, Taipei, APAC.
\n
\n
Responsibilities
\n
\n
\n - Architect and design high-performance, highly available network fabrics for GPU clusters (InfiniBand or RoCE), including topology, cabling, switch configuration, subnet/partitioning, and redundancy.
\n - Plan, design, and implement network infrastructure for GMI global data center, including WAN, Core Network, Data Center Network, Firewalls, Load Balancers, DNS, VPN, etc.
\n - Build high-performance network solutions to support AI/ML workloads, encompassing Compute, Storage, Inband, Management, and Out-of-Band (OOB) network fabric using Infiniband and Ethernet RDMA RoCEv2 technologies.
\n - Lead implementation and lifecycle management of network hardware and firmware (HBAs, NICs, RDMA NICs, IB switches, Ethernet switches, etc).
\n - Configure and fine-tune RDMA, congestion control, QoS, ECN, Priority Flow Control (PFC), and Flow Control to optimize MPI, NCCL, and other GPU-communication patterns.
\n - Hands-on day-to-day operations: provisioning, configuration changes, patching, firmware upgrades, and logging/monitoring of network health.
\n - Rapidly troubleshoot production incidents (fabric errors, performance degradations, link flaps, MTU/flow issues, packet drops), drive root cause analysis, and implement corrective actions.
\n - Develop and maintain automation, scripts, and runbooks for deployment, configuration management, diagnostics, and capacity planning (Ansible, REST APIs).
\n - Work with compute, storage, platform and SRE engineers to validate end-to-end performance, run benchmarks, and recommend architecture improvements; collaborate with cross-functional teams and stakeholders to understand networking requirements.
\n - Identify suitable network providers, vendors, and solutions to meet organizational needs.
\n - Define and enforce network security, segmentation, and access policies for GPU clusters and management networks.
\n - Maintain documentation: network diagrams, cabling maps, configuration baselines, and operational procedures.
\n - Mentor junior network engineers and participate in on-call rotations for fabric support.
\n - Regional/international travel to GMI data center locations.
\n
\n
\n
Meeting every qualification is not required—if you’re excited about this role, we’d love to hear from you. We believe diverse perspectives and experiences strengthen our team.
\n
Qualifications
\n
\n - Bachelor’s degree in Computer Science or related field.
\n - 10+ years of networking experience, with 3+ years specifically designing and operating InfiniBand/RoCE-based fabrics for large-scale GPU AI clusters (hundreds to thousands of GPUs).
\n - Deep, hands-on experience with InfiniBand (HDR/XDR) fabrics: subnet manager (UFM, OpenSM), IB routing, partition keys (P_Key), etc.
\n - Strong expertise with RDMA, RoCE, and Ethernet lossless frameworks: configuring PFC, ECN, DCB, and addressing head-of-line/blocking issues.
\n - Proven troubleshooting experience: packet capture analysis, IB/Ethernet counters, link diagnostics, congestion root-cause, firmware and driver interactions.
\n - Familiarity with GPU communication libraries and patterns: NCCL, MPI, GPUDirect RDMA, and how network settings affect scaling and latency.
\n - Familiarity with storage protocols used in AI environments (NVMe-oF, NFS over RDMA, GPUDirect Storage).
\n - Hands-on with network hardware from major vendors (Mellanox/NVIDIA Spectrum, Cisco, etc).
\n - Solid understanding of TCP/IP, VLANs, L3 routing, BGP/OSPF basics, MTU/Jumbo frames, and L2/L3 troubleshooting tools.
\n - Experience with monitoring and observability tools (Prometheus, Grafana, SNMP, ELK, vendor telemetry).
\n - Experience with network automation and scripting (Ansible, REST API), and configuration management.
\n - Familiar with various routers, switches, firewalls, load balancer, DNS, VPN configuration implementation.
\n - Strong knowledge in network security, DDOS, IDS, etc.
\n - Familiar with optical networking, including fibers, transceivers and optics troubleshooting.
\n - Candidates holding network certifications (e.g. CCNA, CCNP, Nvidia) will be strongly preferred.
\n - Candidates with proven experience in the AI/ML GPU networking environment will be highly considered.
\n - Strong troubleshooting mindset: methodical, data-driven, and calm under production pressure.
\n - Proactive about automation, reliability, and continuous improvement.
\n - Bilingual English and Chinese will be strongly preferred.
\n - Collaborative team player, able to work cross-functionally and to translate technical trade-offs for stakeholders with strong communication skills.
\n
\n
Job Tags