Technical Lead - GPU Infrastructure (100% Remote - Worldwide)

About the Job

Responsibilities

• Own end-to-end platform architecture through proposals, high-level and low-level designs, reviews, and maintenance of the architecture baseline.
• Lead and line-manage backend, frontend, DevOps, QA, and documentation engineers across Europe and India.
• Design, build, and operate a managed Slurm service covering controllers, accounting, partitions, login nodes, onboarding, upgrades, health detection, storage, identity, and isolation.
• Own Kubernetes bootstrap and lifecycle on partner bare metal, GPU operators, VM-based GPU isolation, upgrades, backup, recovery, and node replacement.
• Own managed inference serving, multi-GPU and multi-node parallelism, autoscaling, request routing, endpoint reliability, and confidential-compute-capable capacity.
• Develop metrics, logging, alerting, SLOs, incident response, post-incident reviews, and a sustainable on-call model.
• Interface with infrastructure partners and vendors on specifications, acceptance tests, escalations, capacity planning, and hardware sourcing.
• Translate research, model-training, and product workloads into platform requirements and broker capacity when necessary.
• Complete the platform team and establish technical standards for new engineers.

Must Have

• Eight or more years of hands-on engineering, including at least three years leading infrastructure platform teams.
• Bachelor’s or Master’s degree in computer science or engineering, or equivalent practical experience.
• Hands-on Slurm operations for real users, including scheduling, accounting, node health, and upgrades.
• Bare-metal NVIDIA GPU fleet operations covering drivers, CUDA, Fabric Manager, NVSwitch, DCGM, MIG, and node acceptance.
• InfiniBand, RDMA, SR-IOV, and multi-node NCCL troubleshooting experience.
• Deep Linux systems experience with drivers, passthrough, vfio-pci, cgroups, namespaces, and performance tuning.
• Production Kubernetes operations covering control planes, upgrades, networking, storage, operators, controllers, and multi-tenancy.
• HPC storage and data movement experience with shared filesystems, local NVMe caching, and distributed models and datasets.
• Observability and operations experience with Prometheus, Grafana, Loki or equivalents, SLOs, incident response, and post-incident reviews.
• Working fluency in JavaScript and Node.js for reviewing control-plane, CLI, and worker services.
• A shipped multi-tenant IaaS, PaaS, or research computing service with isolation, quotas, metering, and user-facing APIs or CLIs.
• Excellent written and spoken English, remote availability between UTC and UTC+5:30, and occasional travel.

Desirable

• Experience with Slurm operators on Kubernetes or Kubernetes-native schedulers such as Kueue, Volcano, KAI, or Kubeflow Trainer.
• Experience with modern serving stacks such as vLLM, SGLang, or TensorRT-LLM.
• Experience with GPU workload isolation using KubeVirt, Kata Containers, QEMU, KVM, Firecracker, or confidential-computing technologies.
• Experience with Cluster API, kubeadm, Cilium, autohealing, infrastructure as code, or GitOps.
• Experience in a GPU cloud, national or university HPC centre, AI lab platform team, peer-to-peer systems, or distributed systems.

Workplace

• The role is fully remote and based between UTC and UTC+5:30 so the working day overlaps with Europe and India.
• Occasional travel to partner sites and team events is expected.