Backend Platform Engineer – GPU Cluster Automation

TensorWave; Las Vegas (NV); On-site; USD 140,000 - 210,000; Full time

Benefits offered by this job

Stock Options

Excellent health insurance

401(k)

Flexible PTO

Parental Leave

Life Insurance

Disability Insurance

Employee Assistance Program

Pet & Legal Insurance

Job summary

TensorWave, a provider focusing on AI compute at scale in the United States, seeks a Backend-focused Software Engineer to automate provisioning and operation of large GPU clusters across bare metal, Kubernetes, and Slurm. You will build tooling, pipelines, and health checks to enable reliable, scalable GPU node operations.

The role emphasizes production Go, Kubernetes internals, and observability with Prometheus, Grafana, and OpenTelemetry, contributing to a robust, secure cloud platform.

Qualifications

• 5+ years in infrastructure/platform engineering.
• 3+ years writing production Go.
• Deep understanding of Kubernetes internals: Informers and work queues; Controller-runtime and client-go; CRDs, custom controllers, and operators; Admission webhooks.
• Experience building Kubernetes Operators.
• Experience building gRPC and REST APIs in Go at production scale.
• Familiarity with bare metal concepts (PXE, IPMI, BMC).
• Strong testing discipline across unit, integration, and end-to-end tests.
• Proven ownership of observability stacks such as Prometheus, Grafana, OpenTelemetry, Loki.

Responsibilities

• Build and maintain automated pipelines for provisioning GPU clusters from zero to production.
• Automate Slurm and Kubernetes cluster lifecycle, including bootstrapping, upgrades, provisioning, and decommissioning at scale.
• Develop and maintain infrastructure for GPU node configuration (drivers/firmware).
• Own cluster validation pipelines, automated health checks, and GPU burn-in tests.
• Implement day-2 operations automation: node remediation, rolling upgrades, and automated drain/cordon workflows.
• Write and maintain runbooks and docs for reliable operations.
• Own the full observability stack for automation services and cluster health.

Skills

Go

Kubernetes

gRPC

REST APIs

Observability

CI/CD

Automation

Linux

Networking

Tools

Prometheus

Grafana

OpenTelemetry

Loki

GitHub Actions

Argo CD

Terraform

Ansible

Job description

TensorWave, a provider focusing on AI compute at scale in the United States, seeks a Backend-focused Software Engineer to automate provisioning and operation of large GPU clusters across bare metal, Kubernetes, and Slurm. You will build tooling, pipelines, and health checks to enable reliable, scalable GPU node operations.

The role emphasizes production Go, Kubernetes internals, and observability with Prometheus, Grafana, and OpenTelemetry, contributing to a robust, secure cloud platform.