Principal Engineer, Cluster Orchestration

CoreWeave runs some of the largest GPU clusters in the world. The AI infrastructure behind those clusters determines how workloads are placed, how resources are shared, and how reliably systems perform under constant pressure.

As a Principal Engineer in AI Infrastructure, you will lead the design and evolution of the cluster orchestration systems that make this possible. This includes Slurm, Kubernetes, SUNK, and the control planes that support AI training, inference, and model onboarding at scale.

You will define long-term architecture, solve hard scaling problems, and set technical direction across teams. Your work will directly affect how quickly customers can run models, how efficiently we use GPUs, and how reliably the platform behaves at scale.

• Define the long-term architecture for CoreWeave’s orchestration platforms across Kubernetes, Slurm, SUNK, Kueue, and related systems.
• Act as a technical authority on scheduling, quota enforcement, fairness, pre-emption, and multi-tenant GPU isolation.
• Make design decisions that balance performance, reliability, cost, and operational complexity.

• Lead the evolution of Kubernetes-native control planes, including SUNK and custom operators.
• Design systems that support workload admission, validation, and rollout, including model onboarding flows.
• Identify and remove scaling limits across schedulers, control planes, registries, networking, and storage.

• Set standards for reliability, observability, and operational readiness across orchestration services.
• Define SLOs, alerting, and incident response practices for platform-critical systems.
• Ensure systems behave predictably during failures, peak load, and rapid growth.

• Write and review production code for Kubernetes controllers, schedulers, admission logic, and internal tooling.
• Measure and improve scheduling latency, container startup time, image distribution, and cold-start performance.
• Lead architecture and design reviews across infrastructure teams.

• Mentor senior and staff engineers and help grow technical leaders.
• Influence platform, infrastructure, security, and product teams through clear technical judgment.
• Engage with customers and open-source communities on deep technical topics when needed.

• 15+ years of experience building and operating large-scale distributed systems.
• Deep, practical knowledge of Kubernetes and Slurm internals.
• Experience running GPU-heavy platforms for AI training, inference, or HPC workloads.
• Strong background in Go and cloud-native systems development.
• Proven ability to set technical direction across teams without direct authority.
• Comfortable making high-impact technical decisions in complex systems.
• Bachelor’s or Master’s degree in a relevant field, or equivalent experience.

• Experience with systems such as Kueue, Kubeflow, Argo Workflows, Ray, Istio, or Knative.
• Background in ML platform engineering, model onboarding, or lifecycle management.
• Strong understanding of scheduling strategies, pre-emption, quota enforcement, and elastic scaling.
• Track record of operating highly reliable systems with clear SLOs and incident processes.
• Contributions to Kubernetes, ML infrastructure, or related open-source projects.
• Experience mentoring senior engineers and raising engineering standards.

You may be a good fit if you enjoy defining long-term architecture, solving deep systems problems, and working close to the hardware layer of AI platforms. This role suits engineers who care about correctness, scale, and operational discipline, and who want their work to directly shape how AI runs in production.

At CoreWeave, AI infrastructure is the product. As a Principal Engineer in cluster orchestration, you will be responsible for systems that directly determine how efficiently GPUs are used, how reliably large models run, and how quickly customers can move from research to production.

This role puts you at the center of hard problems in scheduling, resource isolation, and large-scale control planes. You will work on systems where small design choices affect thousands of GPUs and real customer workloads. If you care about building infrastructure that runs under constant pressure, scales without shortcuts, and enables the next generation of AI workloads, CoreWeave is a place where your work will matter.

While we prioritize a hybrid work environment, remote work may be considered for candidates located more than 30 miles from an office, based on role requirements for specialized skill sets. New hires will be invited to attend onboarding at one of our hubs within their first month. Teams also gather quarterly to support collaboration.

California Consumer Privacy Act - California applicants only

CoreWeave is an equal opportunity employer, committed to fostering an inclusive and supportive workplace. All qualified applicants and candidates will receive consideration for employment without regard to race, color, religion, sex, disability, age, sexual orientation, gender identity, national origin, veteran status, or genetic information.

As part of this commitment and consistent with the Americans with Disabilities Act (ADA), CoreWeave will ensure that qualified applicants and candidates with disabilities are provided reasonable accommodations for the hiring process, unless such accommodation would cause an undue hardship. If reasonable accommodation is needed, please contact: careers@coreweave.com.

This position requires access to export controlled information.  To conform to U.S. Government export regulations applicable to that information, applicant must either be (A) a U.S. person, defined as a (i) U.S. citizen or national, (ii) U.S. lawful permanent resident (green card holder), (iii) refugee under 8 U.S.C. § 1157, or (iv) asylee under 8 U.S.C. § 1158, (B) eligible to access the export controlled information without a required export authorization, or (C) eligible and reasonably likely to obtain the required export authorization from the applicable U.S. government agency.  CoreWeave may, for legitimate business reasons, decline to pursue any export licensing process.