3 of 3 InfiniBand Jobs in Central London

highly desirable A strong troubleshooting mindset with the ability to understand behaviour at cluster, hardware and network level Nice to have Exposure to InfiniBand , bare-metal provisioning or HPC-style networking Experience supporting training or inference environments for large-scale ML models Background in AI infrastructure start-ups, hyperscalers ...

routing issue, congestion event or weak fabric design can directly impact customer training workloads. This person will own network architecture across GPU fabric, InfiniBand, RoCE v2, Ethernet leaf spine, edge connectivity, peering, observability, deployment standards and operational handover. We are looking for someone who has: Deep GPU cluster … deployment experience Strong InfiniBand production experience RoCE v2 experience at scale Ethernet fabric experience across BGP, ECMP and low latency operations IPv6 and public ASN experience Multi vendor experience across environments such as NVIDIA Mellanox, Arista, Juniper, Cisco, whitebox or ODM hardware Experience from a neo cloud, hyperscaler, major vendor ...

High-Level Designs aligned with NVIDIA reference architectures and point-to-point deployment standards Design cable tray layouts, basket pathways, and segregation strategies for InfiniBand, Ethernet, out-of-band management, and fibre infrastructure Conduct remote and on-site colocation assessments covering power availability, cooling capacity, floor loading, tray utilization … years of experience in data center infrastructure, HPC, networking, colocation, OEM, hyperscaler, or systems integration environments Practical experience with high-density GPU deployments, InfiniBand/Ethernet networking, structured cabling, and liquid-cooled rack environments Strong understanding of critical infrastructure concepts including: PDUs (single-phase and three-phase) Fibre standards ...