Senior SRE: AI Cloud Reliability & GPU Scale

Nebius Group; Greater London; On-site; GBP 90,000 - 150,000; Full time

Benefits offered by this job

Competitive pay

Career growth

Flexibility

Collaborative culture

Impactful AI projects

International team

Job summary

Nebius is seeking an experienced Site Reliability Engineer to own the reliability, performance, and observability of the full inference stack. You will design telemetry pipelines, tune Kubernetes autoscalers, and craft Terraform modules to ensure cost efficiency and resilience.

You’ll respond to incidents, drive post-morts, and collaborate with software engineers to turn reliability into a product feature. This role requires deep experience with Kubernetes, Prometheus, Grafana, Terraform, and

Qualifications

• Deep fluency with Kubernetes, Prometheus, Grafana, and Terraform.
• Proficiency in Python or Bash scripting.
• Production experience with GPU-heavy workloads (vLLM, Triton, Ray).
• Background in MLOps or model-hosting platforms; focus on reliability.

Responsibilities

• Own reliability, performance, and observability of the inference stack.
• Design telemetry pipelines—metrics, logs, traces—to drive insight.
• Tune Kubernetes autoscalers to improve GPU efficiency and cost.
• Craft Terraform modules to bake resilience into new clusters.
• Harden request-routing and retry logic and drive post-mortems to prevent recurrence.

Skills

SRE practices

Distributed systems

Observability

Incident response

Tools

Kubernetes

Prometheus

Grafana

Terraform

Python

Bash

vLLM

Triton

Ray

Job description

Nebius is seeking an experienced Site Reliability Engineer to own the reliability, performance, and observability of the full inference stack. You will design telemetry pipelines, tune Kubernetes autoscalers, and craft Terraform modules to ensure cost efficiency and resilience.

You’ll respond to incidents, drive post-morts, and collaborate with software engineers to turn reliability into a product feature. This role requires deep experience with Kubernetes, Prometheus, Grafana, Terraform, and