AI Enablement Engineer – NVIDIA AI Enterprise & Kubernetes

Ampstek; Stockholms kommun; Hybrid; SEK 900,000 - 1,300,000; Full time

Job summary

Ampstek is seeking an infrastructure-focused engineer to deploy, configure and validate an AI software stack on Kubernetes for a private Enterprise AI platform running on VMware and NVIDIA GPUs in Stockholm. The role is hybrid with 2 days per week in office.

You will deploy NVIDIA AI Enterprise on Kubernetes, enable GPU sharing, expose metrics with Prometheus and Grafana, validate components on the VKS platform, and run scalability tests for GPU and storage clusters.

Qualifications

• Experience with NVIDIA AI Enterprise on Kubernetes and GPU utilization.
• Experience deploying and validating AI software stacks on Kubernetes.
• Strong knowledge of Prometheus, Grafana, and DCGM exporter for monitoring.
• Hands-on experience with LLM inference serving (NIM, Triton, or vLLM).
• Active security clearance.
• Fluent English, written and spoken.

Responsibilities

• Deploy and configure VMware Private AI Foundation with NVIDIA and NVIDIA AI Enterprise components, including NIM microservices and NeMo Guardrails
• Configure GPU sharing (MIG and/or vGPU time-slicing) for efficient GPU use across workloads
• Expose GPU and LLM workload metrics to Prometheus (e.g. DCGM exporter) with Grafana dashboards
• Validate AI components on the Kubernetes (VKS) platform
• Run and document scalability tests for GPU and storage clusters

Skills

Fluent English

Security clearance

Tools

NVIDIA AI Enterprise on Kubernetes

GPU sharing with MIG or vGPU

Kubernetes in production

Helm

Prometheus & Grafana monitoring

DCGM exporter

LLM inference serving (NIM, Triton or vLLM)

Job description

Location: Stockholm, Sweden (Hybrid – 2 days per week in office)

Contract: 12+ months

Ampstek is hiring for a large enterprise client building a private AI platform on VMware and NVIDIA GPUs. You will deploy, configure and validate the AI software stack that internal teams use to run LLM workloads on Kubernetes.

What you will do

• Deploy and configure VMware Private AI Foundation with NVIDIA and NVIDIA AI Enterprise components, including NIM microservices and NeMo Guardrails
• Configure GPU sharing (MIG and/or vGPU time-slicing) for efficient GPU use across workloads
• Expose GPU and LLM workload metrics to Prometheus (e.g. DCGM exporter) with Grafana dashboards
• Validate AI components on the Kubernetes (VKS) platform
• Run and document scalability tests for GPU and storage clusters

Must have

• NVIDIA AI Enterprise on Kubernetes, including the NVIDIA GPU Operator and NIM deployment
• GPU sharing with MIG and/or vGPU time-slicing
• Kubernetes in production, including Helm
• GPU monitoring with Prometheus and Grafana (incl. DCGM exporter)
• Hands-on LLM inference serving and performance testing (NIM, Triton or vLLM)
• Active security clearance
• Fluent English, written and spoken

Nice to have

• VMware Private AI Foundation with NVIDIA
• VKS or vSphere with Tanzu

Please note: this is an infrastructure role. Data scientists / ML researchers focused on model development, and Kubernetes engineers without GPU experience, will not be a fit.