← All roles

GPU Cluster Engineer, Systems & Platform

Sciforium· Engineering
Apply on company site
San Francisco, CA$150k - $220k + equityFull-timePosted 2mo ago

About the role

GPU Cluster Engineer responsible for the entire software stack of Sciforium's GPU clusters, from kernel tuning and drivers to schedulers, containers, and ML frameworks. Owns production-ready node software, images, pipelines, and fleet consistency to support foundation model training and model serving/product teams.

Responsibilities

  • OS Bring-Up & Node Lifecycle Engineering: Golden Images & Automated Bring-Up, Validation & Burn-In, Fleet Maintenance, Self-Healing Operations
  • Configuration Management & Automation: Infrastructure as Code with Ansible/SaltStack, Provisioning Pipelines, Operational Tooling
  • Orchestration & Scheduling (Kubernetes & Slurm): Kubernetes for Serving, Training Schedulers, Container Platform maintenance
  • GPU Driver & ML Stack Engineering: Driver & Runtime Lifecycle for NVIDIA/AMD stacks, Framework Environments, Distributed Performance tuning
  • Advanced Debugging & Observability: NCCL/RCCL debugging, Observability with DCGM/Prometheus/Grafana, performance reporting

Qualifications

  • 5+ years in systems/infrastructure engineering with GPU cluster/HPC/large-scale ML infrastructure experience
  • Bachelor's or Master's in CS/CE/EE or related field
  • Deep Linux internals expertise (kernel modules/DKMS, systemd, cgroups, NUMA, performance tuning)
  • Hands-on experience with NVIDIA CUDA and/or AMD ROCm driver/runtime stacks on modern accelerators, including kernel debugging
  • Production Kubernetes with GPU workloads; HPC schedulers (Slurm/Run:AI) knowledge
  • Strong configuration management experience (Ansible or SaltStack) with Git-based workflows
  • Provisioning/image tooling experience (Packer, MaaS, Foreman, Terraform) for automated builds
  • Experience with distributed filesystems (Lustre, GPFS, Weka) and checkpoint I/O optimization
  • Container knowledge (Docker/containerd, NVIDIA Container Toolkit or ROCm stack)
  • Proficiency in Python and Bash for automation
  • Knowledge of NCCL/RDMA networking (InfiniBand/RoCE, GPUDirect) and PyTorch/JAX runtime behavior

Skills

Required:OS image managementValidation & burn-in automationConfiguration managementProvisioning toolingKubernetes & GPU schedulingSlurm/Run:AIDriver & runtime engineeringPython & Bash scripting

Benefits

  • Medical, dental, and vision insurance
  • 401k plan
  • Daily lunch, snacks, and beverages
  • Flexible time off
  • Competitive salary and equity
Last updated Sep 25, 2026. Interested? Apply on the company site.

About Sciforium

Byte-Native AI Models & Infrastructure

Sciforium is pioneering the future of AI infrastructure and research. We're developing byte-native multimodal foundation models while delivering serverless LLM serving at a fraction of traditional costs

Company profileWebsite

More open roles at Sciforium