Senior InfiniBand/GPU Cloud Networking Architect

Bitdeer (NASDAQ: BTDR); Penang; On-site; MYR 180,000 - 240,000; Full time

Benefits offered by this job

Training & mentoring

Open workspaces

Welfare benefits

Job summary

Bitdeer, a world‑leading technology company in Bitcoin mining and AI cloud, seeks an experienced Data Center Networking Engineer in Penang. The role focuses on InfiniBand/RoCE fabrics, GPU cluster networking at scale, and advanced fabric management using UFM tools.

You will diagnose, tune, and automate network operations, collaborating with Nvidia/Mellanox support, and driving telemetry‑driven improvements for reliability and performance.

Qualifications

• 5+ years in data center networking, with at least 3 years on InfiniBand or RoCE fabrics.
• Hands-on experience deploying and operating Nvidia/Mellanox InfiniBand switches at scale.
• Strong understanding of IB subnet management, partitioning, and QoS.
• Experience with RoCEv2 deployment including PFC, ECN, and DCQCN configuration.
• Proficiency with UFM or equivalent IB fabric management tools.
• Knowledge of 400G/800G optics, cabling standards and structured cabling.
• Experience diagnosing IB/RoCE issues using ibdiagnet, perfquery, ibstat.
• Understanding of NCCL and GPU communication mapping to network topology.
• Telemetry-driven ops mindset; ability to build dashboards/alerts on RDMA counters.
• Runbook-as-code mindset; turn diagnostics into automation.

Responsibilities

• Manage InfiniBand/RoCE fabrics and QoS for GPU clusters of 100–10,000 GPUs.
• Coordinate with Nvidia/Mellanox support for escalations and fixes.
• Monitor IB and RoCE performance; tune routing and congestion control.
• Tune NCCL topology and GPU communication mapping to network topology.
• Instrument telemetry into the data platform and define predictors for reliability.
• Develop incident workflows to automate remediation and recovery.

Skills

InfiniBand

RoCE fabrics

Nvidia/Mellanox IB switches

IB subnet management

QoS

RoCEv2 deployment

PFC

ECN

DCQCN

UFM tooling

400G/800G optics

NCCL knowledge

Telemetry-driven ops

Tools

ibdiagnet

perfquery

ibstat

PortRcvErrors

PortXmitDiscards

Job description

Bitdeer, a world‑leading technology company in Bitcoin mining and AI cloud, seeks an experienced Data Center Networking Engineer in Penang. The role focuses on InfiniBand/RoCE fabrics, GPU cluster networking at scale, and advanced fabric management using UFM tools.

You will diagnose, tune, and automate network operations, collaborating with Nvidia/Mellanox support, and driving telemetry‑driven improvements for reliability and performance.