ML Infra Engineer: Scheduling & Kubernetes

Sponsor Finder; Greater London; Hybrid; GBP 110,000 - 170,000; Full time

Job summary

Isomorphic Labs Limited is seeking a Software Engineer (Scheduler) in London to lead the scheduling layer for ML workloads. You will design and operate scheduling systems, optimize resource use across GPUs/TPUs, and build custom Kubernetes controllers to support our ML pipelines.

You will work closely with ML researchers and the Infrastructure team to deliver reliable, scalable compute orchestration. Strong Python/Go and Kubernetes expertise are essential.

Qualifications

• Own the Scheduling Layer: design, operate, and extend ML workload scheduling systems.
• Optimize compute utilization with gang scheduling and topology-aware placement.
• Partner with ML researchers and Infrastructure to turn requirements into platform capabilities.
• Build reliable systems with clear failure semantics and defined SLOs.
• Develop custom Kubernetes controllers/CRDs to tailor the cluster for ML pipelines.
• Experience with cluster schedulers e.g., Kueue, Volcano, Slurm and Kubernetes internals.
• Strong cloud infrastructure experience and Infrastructure as Code practices.
• Proficiency in Python or Go for tooling and automation.

Responsibilities

• Own and evolve the scheduling layer to manage dynamic batch inference and long-running training workloads.
• Drive scheduling policies to maximize GPU/TPU utilization across fleets.
• Collaborate with researchers to understand workload patterns and translate needs into platform features.
• Ensure high availability through robust runbooks and reliable operational processes.
• Create and maintain custom Kubernetes controllers and integration points.

Skills

Scheduling systems

Kubernetes

Python/Go

DevOps/SRE

Cloud infrastructure

Tools

Kubernetes controllers

CRDs

K8s scheduler

Volcano / Slurm

Infrastructure as Code

Job description

Isomorphic Labs Limited is seeking a Software Engineer (Scheduler) in London to lead the scheduling layer for ML workloads. You will design and operate scheduling systems, optimize resource use across GPUs/TPUs, and build custom Kubernetes controllers to support our ML pipelines.

You will work closely with ML researchers and the Infrastructure team to deliver reliable, scalable compute orchestration. Strong Python/Go and Kubernetes expertise are essential.