Infrastructure Engineer (ML Systems) - London
Hybrid
Core
Own GPU compute infrastructure for model training, evaluation, and serving systems.
Role type
Senior Infrastructure Engineer (ML Systems)
Builds
GPU clusters, schedulers, and capacity management for research and product teams
Domain
Cloud Infrastructure / Machine Learning Systems
Deliverable
infrastructure
Required skills
Kubernetes, GPU cluster design, Linux OS debugging, Python, GPU drivers, networking, storage
Preferred skills
SRE experience, Slurm, Kueue, Karpenter, SkyPilot, AWS EKS/SageMaker, InfiniBand, NCCL, Terraform
Responsibilities
Run Kubernetes clusters for GPU training and inference, improve GPU scheduling and placement, detect and repair unhealthy GPU nodes, manage cloud capacity, debug failures in GPUs/Linux/storage/networking
Seniority
Senior, hands-on IC
