Principal AI Infrastructure Engineer — GPU Fleet

Nscale; Bellevue (WA); On-site; USD 240,000 - 400,000; Full time

Benefits offered by this job

Medical, dental, vision

Flexible paid time off

Parental leave

Retirement plan

Job summary

Nscale is building a GenAI cloud from sustainable data centers and advanced AI infrastructure. The Principal Software Engineer will lead the Fleet Manager platform, owning architecture, design, and delivery of Python-based automation systems that provision, test, and remediate GPU nodes and network switches at scale.

You will mentor senior engineers, set technical direction, ensure reliability and observability, and partner with Infrastructure, Platform, and SRE teams to deliver production-grade

Qualifications

• 12–15+ years of production systems experience with leadership in infra automation or workflow tooling.
• Strong Python fundamentals and experience leading multi-service distributed systems.
• Experience delivering large-scale automation and infrastructure projects from requirements to production.
• Familiarity with AI tooling as part of development workflows (e.g., Claude, Cursor).
• Proven ability to drive reliability, observability, and SRE practices.

Responsibilities

• Lead architecture and roadmap for Fleet Manager workflow automation systems.
• Own end-to-end device provisioning, testing, and remediation workflows at scale.
• Design and build workflow orchestration for GPU and network lifecycle management.
• Set engineering standards for reliability, monitoring, and operational excellence.
• Mentor senior engineers and drive technical direction across teams.
• Collaborate with Infrastructure, Platform, and SRE to translate needs into scalable automation.

Skills

Leadership

Python

Distributed systems

AI tooling

SRE practices

Mentorship

Tools

Temporal

Airflow

Prefect

Kubernetes

Terraform

Pulumi

MAAS

Ironic

IPMI

NetBox

OpenStack

Job description

Nscale is building a GenAI cloud from sustainable data centers and advanced AI infrastructure. The Principal Software Engineer will lead the Fleet Manager platform, owning architecture, design, and delivery of Python-based automation systems that provision, test, and remediate GPU nodes and network switches at scale.

You will mentor senior engineers, set technical direction, ensure reliability and observability, and partner with Infrastructure, Platform, and SRE teams to deliver production-grade