GPU Fabric Network Engineer – RoCE & InfiniBand

Kyndryl Inc.

Jakarta Pusat

On-site

IDR 300,000,000 - 600,000,000

Full time

2 days ago

Be an early applicant

Application generator

An application made for this job — a tailored resume and cover letter that speak straight to the posting.

Get past ATS filters

Job summary

Kyndryl in Jakarta is seeking an RoCE / InfiniBand Network Engineer to provide hands-on troubleshooting and change execution for high-speed GPU fabrics. You will work with RoCE and InfiniBand environments, own service restoration, and escalate complex fabric issues to SMEs while contributing to runbooks and SLA adherence.

This role offers a hybrid-friendly culture, opportunities to grow into senior fabric specialist, architecture or leadership roles, and exposure to RDMA, NCCL workloads, and

Qualifications

3-5+ years of data-center network engineering with high-speed fabric.
RoCEv2/RDMA and/or InfiniBand; GPU/HPC environments preferred.
Switch CLI, optics/cabling, NIC/HCA diagnostics and counters.
Knowledge of DCB, PFC, ECN, QoS and lossless Ethernet for RoCE.
Leaf-spine, BGP/EVPN and 100/200/400 Gb fabrics experience.
Formal incident, change management and RCA exposure.

Responsibilities

Troubleshoot switch, HCA/NIC, port and optic issues via CLI and telemetry.
Analyze link errors, drops, flaps, speed and RDMA reachability.
Operate InfiniBand/UFM or Spectrum-X/RoCE tooling within policy.
Triage RoCE and InfiniBand fabric changes per site architecture.
Own SLA progression and contribute to RCA and runbooks.

Skills

RoCEv2

RDMA

InfiniBand

GPU HPC

BGP

EVPN

Spectrum-X

ConnectX

UFM

DCB

PFC

QoS

Telemetry

Fabric-tools

Tools

NVIDIA Spectrum-X

ConnectX

UFM

Job description

This role offers a hybrid-friendly culture, opportunities to grow into senior fabric specialist, architecture or leadership roles, and exposure to RDMA, NCCL workloads, and

Get your free, confidential resume review.

or drag and drop your file here.