Senior GPU Cluster Orchestration Engineer

Verda; Helsinki; Remote; EUR 90,000 - 170,000; Full time

Benefits offered by this job

Cash and equity

Global team

Impactful work

Job summary

Verda is hiring a Senior Cluster Orchestration Engineer to build the shared GPU capacity layer that unifies training, inference and batch workloads across Kubernetes and SLURM.

You will write schedulers, controllers and platform services, collaborating with GPU fleet engineers and product teams. The role offers direct architecture influence on state-of-the-art hardware at scale and a path from idea to production.

Qualifications

• 5+ years in platform engineering or related field, with design-to-deploy experience.
• Strong Go or Python development skills and Kubernetes controller experience.
• Deep Kubernetes knowledge: scheduling framework, control plane, networking and storage.
• Hands-on with batch or AI schedulers like Volcano, KAI Scheduler, Kueue or Slurm Bridge.

Responsibilities

• Design and build the scheduling layer to share GPU capacity across Kubernetes and SLURM.
• Add support for perishable compute and short-lived, preemptible workloads.
• Implement gang scheduling and topology-aware placement for multi-node training.
• Develop Kubernetes operators for quotas, priorities, and queueing; manage cluster lifecycle.

Skills

Go

Python

Kubernetes

API design

Concurrency

Distributed systems

Tools

Volcano

KAI Scheduler

Kueue

Slurm Bridge

Job description

Verda is hiring a Senior Cluster Orchestration Engineer to build the shared GPU capacity layer that unifies training, inference and batch workloads across Kubernetes and SLURM.

You will write schedulers, controllers and platform services, collaborating with GPU fleet engineers and product teams. The role offers direct architecture influence on state-of-the-art hardware at scale and a path from idea to production.