Senior DevOps Engineer (Kubernetes / GPU / AI Infrastructure)

Senior DevOps Engineer for a metallurgical enterprise. Remote from Russia, full-time, salary from 200,000 RUB/month. Kubernetes, NVIDIA GPU, ML/AI infrastructure.

About the project: large metallurgical enterprise. Tasks: deploying and maintaining Kubernetes clusters on bare metal; configuring and optimizing Docker; developing GitLab CI/CD pipelines; managing NVIDIA GPU and CUDA infrastructure; deploying and supporting ML inference services (vLLM, Ollama, Triton Inference Server); setting up monitoring with Prometheus and Grafana; working with network services, TLS and DNS; managing Kubernetes deployments with Helm; administering GitLab Container Registry.

Required: confident Linux administration (Ubuntu Server); practical experience with Docker and Kubernetes; experience with Helm; confident GitLab CI/CD and GitLab Container Registry; experience with NVIDIA GPU and CUDA; knowledge of NVIDIA Container Toolkit and Kubernetes NVIDIA Device Plugin; experience with Prometheus and Grafana monitoring; understanding of networking, TLS and DNS. Nice to have: experience with GPU servers for AI/ML; work with vLLM, Ollama, Triton Inference Server; understanding of CUDA, VRAM, tensor parallelism and GPU specifics; experience with bare metal Kubernetes; familiarity with NVIDIA DCGM Exporter and Langfuse; experience preparing infrastructure for GDPR; experience migrating applications between cloud and on-prem.

Fully remote work from Russia. Full-time employment on a long-term project. Work with modern Kubernetes and GPU infrastructure stack. Monthly payment. Candidates with Russian citizenship considered.