GPU ML Infrastructure Engineer — End-to-End Acceleration

Arbeitnow; Amsterdam; Hybrid; EUR 110,000 - 140,000; Full time

Benefits offered by this job

Hybrid work model

Competitive compensation

MacBook Pro

Career growth

Job summary

Prosus seeks an experienced ML Infrastructure & Acceleration Engineer to own the GPU infrastructure powering LLM training and serving. You will run our H200 cluster end to end, deploy models to production, and optimize pipelines down to the kernel level.

You will balance reliability with cost, enabling rapid research iteration for millions of users. You will collaborate with ML engineers and researchers to establish best practices, monitor system health, and push for continuous improvements in

Qualifications

• Minimum 5+ years in ML infrastructure, MLOps, HPC, or platform engineering with GPU focus.
• Hands-on experience operating multi-node GPU clusters with Kubernetes and/or Slurm, debugging hardware and networking.
• Production experience deploying and optimizing LLM inference with vLLM, SGLang, or TensorRT-LLM.
• Proven track record profiling and accelerating GPU workloads with measurable results.
• Solid monitoring/observability for GPU systems (Prometheus, Grafana, DCGM).
• Curious mindset with a drive to implement improvements.

Responsibilities

• Own GPU cluster end-to-end (H200), using Kubernetes and Slurm for training/inference.
• Maintain healthy distributed runs: node checks, NCCL debugging, auto-recovery, checkpointing.
• Optimize GPU utilization and cost with dashboards and alerts.
• Manage infrastructure-as-code and provisioning for GPU environments.
• Define and monitor SLIs/SLOs for cluster availability and model serving.
• Deploy LLMs to production using vLLM/SGLang with CI/CD and safe rollouts.

Skills

ML infrastructure

GPU workloads

Kubernetes

Slurm

NCCL/InfiniBand

Profiling tools

Triton/CUDA kernels

vLLM/SGLang

PyTorch

Performance optimization

Education

Bachelor’s or higher in CS/EE/ML

Tools

Kubernetes

Slurm

NCCL

TensorRT-LLM

PyTorch

Triton

CUDA

Nsight Systems/Compute

Job description

Prosus seeks an experienced ML Infrastructure & Acceleration Engineer to own the GPU infrastructure powering LLM training and serving. You will run our H200 cluster end to end, deploy models to production, and optimize pipelines down to the kernel level.

You will balance reliability with cost, enabling rapid research iteration for millions of users. You will collaborate with ML engineers and researchers to establish best practices, monitor system health, and push for continuous improvements in