Production AI Platform Engineer (Kubernetes, GPU) — Remote

emagine; Polska; Remote; PLN 306,000 - 481,000; Full time

Job summary

emagine is seeking an experienced DevOps / Platform Engineer to own and operate a production AI platform running LLM workloads. This individual contributor role requires hands-on ownership of platform components, with a focus on reliability, security, and predictable production operation.

The stack includes Kubernetes with Talos Linux and GPU-enabled nodes, NVIDIA NIM and vLLM for model serving. You will work closely with engineers and business stakeholders to improve automation, observability

Qualifications

• Proven track record operating Kubernetes in production.
• Strong Linux administration skills.
• Experience with cluster lifecycle management and production upgrades.
• Ability to troubleshoot complex incidents and perform root cause analysis.
• Fluent spoken and written English.

Responsibilities

• Operate and improve a production-grade Kubernetes platform.
• Take end-to-end ownership of selected platform components.
• Troubleshoot Kubernetes infrastructure and production issues.
• Plan and perform safe rolling upgrades of Kubernetes clusters and platform components.
• Improve automation, observability and operational processes.
• Maintain security and data residency standards.

Skills

Kubernetes in production

Linux administration

Incident resolution

Cluster lifecycle management

Observability

Automation

English

Tools

NVIDIA NIM

vLLM

Talos Linux

CUDA

Kubernetes Operators

Kubernetes controllers

Job description

emagine is seeking an experienced DevOps / Platform Engineer to own and operate a production AI platform running LLM workloads. This individual contributor role requires hands-on ownership of platform components, with a focus on reliability, security, and predictable production operation.

The stack includes Kubernetes with Talos Linux and GPU-enabled nodes, NVIDIA NIM and vLLM for model serving. You will work closely with engineers and business stakeholders to improve automation, observability