Staff Kubernetes GPU Orchestration Engineer

Coreweave; Łódź; On-site; PLN 260,000 - 380,000; Full time

Benefits offered by this job

Living Wage Accredited Employer

Job summary

CoreWeave is seeking a Staff Software Engineer, Cluster Orch (Non SUNK), to lead the Kubernetes-native orchestration strategy for massive AI workloads. You will own the technical vision, architecture, and platform-wide standards, mentoring senior engineers and aligning with cross-team efforts.

You’ll work on distributed systems at scale, leveraging Go, Kubernetes internals, and frameworks like Kueue, Volcano, and Ray to ensure performance, scalability, and fairness across large GPU clusters.

Qualifications

• 8+ years of professional software engineering experience in distributed systems or cloud platforms.
• Strong Go programming skills and experience designing large-scale production systems.
• Deep knowledge of Kubernetes internals and scheduling mechanisms.

Responsibilities

• Define the technical vision and architecture for major portions of CoreWeave's orchestration platform.
• Lead cross-team architecture reviews and establish platform-wide reliability and developer experience standards.
• Mentor senior engineers and drive alignment across infrastructure, CKS, and managed inference teams.
• Own performance, scalability, and fairness improvements across large GPU clusters.

Skills

Go

Distributed systems

Kubernetes internals

Scheduling mechanisms

Education

Bachelor's degree in Computer Science, Engineering, or a related technical field

Tools

Kueue

Volcano

Ray

Job description

CoreWeave is seeking a Staff Software Engineer, Cluster Orch (Non SUNK), to lead the Kubernetes-native orchestration strategy for massive AI workloads. You will own the technical vision, architecture, and platform-wide standards, mentoring senior engineers and aligning with cross-team efforts.

You’ll work on distributed systems at scale, leveraging Go, Kubernetes internals, and frameworks like Kueue, Volcano, and Ray to ensure performance, scalability, and fairness across large GPU clusters.