Platform Engineer - AI Infrastructure
Core
Building the control-plane software (scheduling, scaling, multi-tenancy, serving) for a large multi-vendor GPU fleet to support sovereign AI training and inference workloads.
Role type
Senior Platform Engineer (AI Infrastructure)
Builds
Control-plane services, scheduler integrations, autoscaling controllers, inference-serving platform, RBAC/quota systems, observability tooling, and developer CLI/APIs.
Domain
AI Infrastructure / GPU Computing / Cloud Native
Deliverable
production ML models | infrastructure
Required skills
Kubernetes (controller/internals), Go, Python, GPU constraints (MIG, gang scheduling, topology-aware placement), system design, product mindset
Preferred skills
GPU scheduling systems (Kueue, Volcano), multi-tenant isolation, deep Kubernetes networking (CNI, RDMA), on-premise GPU platforms, open-source contributions
Technologies
Kubernetes, Go, Python, Terraform, Crossplane, CNI, RDMA, SR-IOV
Responsibilities
Design and ship control-plane services and scheduler integrations; build autoscaling controllers for training and serving; implement multi-tenancy, RBAC, and isolation layers; develop networking components and observability tooling; create developer self-service CLI and APIs; manage infrastructure-as-code and cluster provisioning.
Seniority
Senior, hands-on IC