Senior Site Reliability Engineer — Token Factory (Inference Platform)

Nebius is seeking a Senior Site Reliability Engineer — Token Factory (Inference Platform) for an IT role based in Berlin. This senior-level position requires a Bachelor's degree and 5+ years of relevant experience. You will join Nebius Cloud, which runs tens of thousands of GPUs, to build an inference platform that deploys foundation models at massive scale.

Core Responsibilities

• Own the reliability, performance, and observability of the entire inference stack.
• Design and refine telemetry pipelines, including metrics, logs, and traces.
• Tune Kubernetes autoscalers and craft Terraform modules to build resilience into clusters.
• Harden request-routing and retry logic for high-throughput APIs.
• Handle incident detection, isolation, remediation, and drive post-mortem cultures.

Requirements and Skills

Candidates must have a Bachelor's degree and 5+ years of experience. Success in the role requires deep fluency with Kubernetes, Prometheus, Grafana, Terraform, and infrastructure-as-code. You should be comfortable scripting in Python or Bash, understanding SLOs, and managing distributed back-ends in production. Experience with GPU-heavy workloads like vLLM, Triton, or Ray, as well as a background in MLOps or model-hosting platforms, will also serve you well. Applicants must be authorized to work in the country.

Benefits and Culture

Nebius offers market competitive compensation, career growth and learning opportunities, flexibility and ownership, and a collaborative, innovative culture working on impactful AI projects in an international environment. Nebius is an equal opportunity employer committed to an inclusive and diverse workplace.