Sr. GPU Cloud East-West Network Expert (SRE SME)

About the role

Bitdeer is seeking a senior expert to design, operate and optimise east‑west networking for large GPU clusters in its AI‑cloud and Bitcoin‑mining datacenters. You will work on high‑performance InfiniBand and RoCE fabrics that scale from hundreds to tens of thousands of GPUs.

Key responsibilities

• Design, deploy and manage InfiniBand fat‑tree, rail‑optimized and dragonfly topologies for GPU clusters of 100–10,000 GPUs.
• Implement and tune RoCEv2 networks on Nvidia, Arista and Cisco platforms for RDMA workloads.
• Operate Unified Fabric Manager (UFM) for fabric monitoring, diagnostics and subnet management.
• Perform performance tuning – adaptive routing, congestion control (DCQCN/ECN), traffic isolation and NCCL communication optimisation.
• Manage firmware lifecycle for IB switches and HCAs, and coordinate escalations with Nvidia/Mellanox support.
• Collect IB/RoCE telemetry (ibdiagnet, perfquery, ibstat, etc.) and feed it into the AIOps fault‑prediction engine.

Required profile

• 5+ years in data‑center networking, with at least 3 years focused on InfiniBand or RoCE fabrics.
• Hands‑on experience deploying Nvidia/Mellanox InfiniBand switches at scale.
• Strong knowledge of IB subnet management, partitioning, QoS and RoCEv2 configuration (PFC, ECN, DCQCN).
• Proficiency with UFM or equivalent fabric‑management tools.
• Understanding of NCCL, GPU communication patterns and telemetry‑driven operations.

Required skills

• InfiniBand (fat‑tree, rail‑optimized, dragonfly)
• RoCEv2 (PFC, ECN, DCQCN)
• Nvidia, Mellanox, Arista, Cisco networking hardware
• Unified Fabric Manager (UFM)
• NCCL communication tuning
• 400G/800G optics and structured cabling standards
• ibdiagnet, perfquery, ibstat diagnostic tools

What we offer

• Inclusive, diverse culture with open workspaces and start‑up spirit.
• Fast‑growing company providing direct impact on digital‑asset and AI‑cloud industries.
• Opportunities to work on cutting‑edge HPC and Bitcoin‑mining projects worldwide.
• Personal accountability, autonomy, and rapid professional growth.
• Attractive welfare benefits, training and mentoring programmes.