Panthalassa

Senior HPC Platform Engineer

Panthalassa
Apply
1 day ago
Portland, OR, USASenior

Base Salary

$160k - $220k/yr

Responsibilities

  • Define the HPC platform’s service catalog, service levels, roadmap, and user-focused priorities with CFD engineering leads.
  • Build golden paths, job templates, submission tooling, software environment conventions, documentation, and self-service workflows for CFD engineers.
  • Automate cluster configuration and operations with Ansible, Git, CI, Terraform, and GitOps practices.
  • Operate HPC clusters end to end, including Linux nodes, Slurm scheduling, Warewulf provisioning, storage, interconnects, BMC/IPMI management, firmware, and hardware fault diagnosis.
  • Plan and provision new compute capacity, including hardware specifications, rack and network layouts, node validation, and supplier/RMA coordination.
  • Build and maintain scientific software environments with Environment Modules or Lmod, MPI stacks, compiler toolchains, and container-based delivery tools.
  • Operate and validate Omni-Path and InfiniBand fabrics, including subnet management, topology, link health, and MPI performance.
  • Administer ZFS pools, NFS exports, snapshots, scrubbing, backups, capacity planning, and shared simulation storage.
  • Create monitoring, alerting, dashboards, runbooks, and operational procedures for nodes, fabrics, scheduling, storage, and user-facing job diagnostics.
  • Partner with CFD engineers on profiling, NUMA and process placement, MPI tuning, and production HELYX, OpenFOAM-family, and Star-CCM+ workloads.
  • Plan hybrid and burst workflows connecting on-premises HPC with Kubernetes and GKE infrastructure.

Requirements

  • At least 5 years of experience operating production Linux HPC clusters or equivalent scientific or batch compute platforms, with deep Slurm configuration, accounting, and scheduling expertise.
  • Experience taking over inherited clusters, auditing and documenting them, and bringing them under configuration management.
  • Experience with diskless or stateless provisioning using Warewulf, xCAT, or similar tools, including PXE and UEFI troubleshooting.
  • Experience operating Omni-Path or InfiniBand interconnects and validating MPI performance.
  • Proficiency with BMC, IPMI, or Redfish management and remote diagnosis of bare-metal hardware faults.
  • Experience building scientific software stacks with Environment Modules or Lmod, MPI libraries, compilers, Ansible, and Git.
  • Experience operating infrastructure as an internal product, including service levels, self-service tooling, and user-impact-based prioritization.
  • Experience with infrastructure-as-code and GitOps workflows, including Terraform, CI pipelines, and pull-request-driven changes.
  • Familiarity with CFD or other engineering simulation workloads and their scheduler, storage, fabric, and scaling requirements.
  • Experience implementing infrastructure monitoring and alerting, user-facing dashboards, and actionable runbooks.
  • Ability to guide on-site staff through physical hardware work with precise instructions.
  • Strong communication skills with engineers and vendors, plus enthusiasm for renewable energy and mission-driven infrastructure.
  • Preferred experience includes solver profiling, NUMA and process placement, HPC telemetry, ZFS, DRBD, Corosync, Pacemaker, parallel filesystems, CFD packages, FlexLM, RLM, Datadog, Prometheus, Grafana, Kubernetes batch workloads, Kueue, Volcano, Slurm on Kubernetes, CLIs, Open OnDemand, internal APIs, platform engineering, Apptainer, Spack, EasyBuild, and GKE on GCP.
  • Legal authorization to work in the United States and availability to intermittently work longer hours or weekends for critical needs.

Benefits

  • Full-time cash compensation of $160,000-$220,000 plus company equity.
  • Flexible paid time off and flexible working hours with a high degree of autonomy and work-life balance.
  • Company-paid gold-level PPO health insurance for full-time employees, partners, and dependents.
  • Company-paid dental and vision insurance for full-time employees, partners, and dependents.
  • Disability insurance and access to tax-advantaged 401(k), health FSA, and dependent care FSA accounts.
  • Relocation assistance for moving to Portland, Oregon.
  • Portland-based role preferred; exceptional remote candidates may be considered with regular travel to company offices and relevant facilities.

Tech Stack

AnsibleDatadogGitGitHub ActionsGoogle Cloud PlatformGrafanaKubernetesLinuxPrometheusTerraform

Categories

Panthalassa

About Panthalassa

51-200 employees

Panthalassa builds ocean-energy platforms that harvest open-ocean power to run clean compute at sea and produce renewable fuels delivered to shore. The company develops and operates these systems as a commercial provider of energy and supporting technology for compute operators and fuel markets. It is a privately held public benefit corporation headquartered in Portland, Oregon, and is backed by venture capital, philanthropic investors, and university endowments.

Contact me