HPC Engineer (InfiniBand), £600-650 per day inside IR35
HPC Engineer (InfiniBand), £600-650 per day inside IR35
3 months initially, Remote Working
Bench IT require an experienced NetOps Team Lead to focus on HPC & InfiniBand engineering. The role requires specialist High-Performance Computing (HPC) fabric experience to take ownership of network reliability and lead the operations function. This role will be working predominately remotely however, some site attendance in Sheffield will be required approximately 2-4 times per month.
The successful candidate should have the skills included below;
- High-Speed AI Fabrics: Deep experience in InfiniBand (Subnet Manager, OpenSM, link training) and/or RoCEv2 (lossless Ethernet).
- Technical knowledge of: InfiniBand, NVIDIA Quantum, RoCE, Mellanox, NetQ, OpenSM, NOC Lead.
- Network Triage: Proven ability to troubleshoot fabric anomalies-specifically optical errors, BER/FEC, and flapping links.
- Operations Leadership: Experience running a NOC/NetOps function, writing runbooks, setting up incident triage, and managing escalations.
- ITSM Integration: Familiarity with modern service desk workflows (HaloITSM or similar).
HPC Engineer (InfiniBand), £600-650 per day inside IR35
3 months initially, Remote Working