Summary

Senior site reliability engineer for Nebius's Token Factory inference platform. The role owns reliability, performance, observability, and incident response across a Kubernetes-based GPU infrastructure stack supporting multimodal AI workloads. The working arrangement is not specified.

Responsibilities

• Design and refine telemetry pipelines
• Tune Kubernetes autoscalers for GPU efficiency
• Create Terraform modules for resilient clusters
• Harden request routing and retry logic
• Automate incident detection and remediation
• Write runbooks for production incidents
• Drive post-mortems to prevent recurrence