Senior Cluster Orchestration Engineer

1 hour agoSenior London (UK), Helsinki (Finland), or elsewhere in Europe remotely Remote Full Time Engineering Jobs by Verda

Verda

Verda is a Helsinki-headquartered full-stack AI cloud provider offering on-demand GPU compute, clusters, serverless containers, storage, APIs, and related AI infrastructure.

Series BRecently funded

Helsinki, Finland

Funding history

Series B and additional investment ($189M)2026 funding round extension (revised total)Equity and debt financing extension

Investors

Skaala byfounders Nordea 6 Degrees Capital (6DC)Emergence Capital

About Verda

Formerly DataCrunch, Verda operates European AI cloud infrastructure for the full AI lifecycle, from prototyping and distributed model training to scalable inference. It manages data-center infrastructure, hardware, cloud-platform services, and an in-house AI Lab.

View jobs by Verda

Skills

Gitops Go Gpu Kubernetes Kueue Observability Orchestration Python Scheduler Slurm Sre Volcano

Candidate Availability

Remote · Helsinki, London · Required

Required and preferred rules are kept separate and reflect the wording in the original posting.

About the Role

You will build scheduling, controller, and platform services that enable training, inference, and batch workloads to share GPU capacity. You will develop Kubernetes operators, improve workload placement and preemption, build observability, and deliver reliable platform changes through GitOps.

Requirements

5+ years of platform engineering, infrastructure engineering, SRE, or related experience
Software development skills in Go or Python
Experience building production Kubernetes controllers or operators
Deep knowledge of Kubernetes scheduling, control plane, networking, storage, and node lifecycle
Experience with Volcano, KAI Scheduler, Kueue, Slurm Bridge, or similar workload schedulers
Understanding of gang scheduling, queueing, quotas, preemption, and fair-share
Experience running GPU workloads and GPU-aware scheduling
Familiarity with SLURM

Responsibilities

Design and build scheduling for shared GPU capacity across Kubernetes and SLURM
Build support for perishable compute and reclaim capacity for higher-priority workloads
Implement gang scheduling, topology-aware placement, and preemption
Develop Kubernetes operators and controllers for quotas, priorities, fair-share, queueing, and cluster lifecycle management
Solve scheduling interactions between training, inference autoscaling, and batch backfill
Build observability for scheduling, utilisation, and platform-service SLOs
Deliver platform changes through GitOps with declarative configuration, drift detection, and safe recovery

Benefits