Senior HPC Systems Engineer at Parallel Works

United States

About Parallel Works<p>Parallel Works builds and operates ACTIVATE, a control plane for high performance computing and AI. Our customers run large scientific and AI workloads across their own on-premises clusters, Government and commercial cloud, and commercial GPU providers, and ACTIVATE gives them one way in to all of it. The high security boundary is authorized at Impact Level 5, with FIPS validated cryptography and STIG hardening throughout.</p><p>The work reaches most fields that depend on computing at scale: weather and climate forecasting, defense and intelligence programs, aerospace and structural analysis, molecular and materials science, energy, and AI research. A quarter here can include standing up a GPU cluster for one of those communities, federating a laboratory's existing on-premises system with burst capacity it did not have before, and getting a domain code written decades ago to run on current hardware.</p><p>Customer success sets our priorities. We are a small engineering company, so engineers here work directly with the people using the systems and carry a problem from the first report through to the fix. This is what we call mission engineering: understanding what a customer is trying to accomplish and why the computing matters to it.</p>About the role<p>Parallel Works is hiring a Senior HPC Systems Engineer to build and run the clusters behind our defense and research programs. The work covers GPU node bring-up, Slurm configuration, fabric and storage troubleshooting, security hardening, and Tier 3 escalation.</p><p>The computing environments are hybrid. Some clusters are customer owned hardware on site, some run in accredited Government cloud regions, and some are dedicated GPU clusters at commercial providers. On several programs the on-premises systems carry the primary load and cloud takes the overflow. The position is senior: it handles the escalations the rest of the team cannot resolve, and it trains the junior engineers.</p>What you will do<ul><li>Cluster operations: build and operate production Slurm clusters. slurmctld and slurmdbd, partitions and QOS, accounts and fair share, GPU GRES, prolog and epilog, cgroup enforcement.</li><li>Hybrid federation: connect customer owned clusters to the control plane, reconciling their site scheduler, storage, and identity source so accounts and allocations behave the same in every venue.</li><li>On-premises hardware: bare metal provisioning, out of band management, firmware, rack networking, and fault coordination with site staff or vendors.</li><li>GPU and fabric: validate GPU nodes before users arrive. Driver and CUDA stack, DCGM health checks, XID triage, fabric manager and NVLink checks, InfiniBand verification, NCCL tuning.</li><li>Storage and automation: tune parallel and high throughput storage, and write the Ansible, Terraform, and image build pipelines that make a cluster reproducible.</li><li>Security and escalation: STIG hardening, scan remediation, FIPS validated cryptography, security package artifacts, Tier 3 escalations, and a share of the on call rotation.</li></ul> <ul><li>10 or more years operating production Linux systems across more than one distribution family. RHEL, Rocky, or Alma on the Government side and Debian or Ubuntu on the commercial GPU side, since those clusters usually ship Ubuntu. Kernel and network tuning, systemd, cgroups, NUMA.</li><li>Production Slurm administration. You have configured, debugged, and upgraded a scheduler other people depended on.</li><li>At least one parallel or high throughput filesystem in production, plus InfiniBand or RoCE fabric operations.</li><li>NVIDIA GPU node operations at multi-node scale, including driver stack management and fault triage.</li><li>Experience on customer owned or on-premises clusters as well as public cloud, including bare metal provisioning and out of band management.</li><li>DevOps experience with infrastructure as code frameworks such as Ansible or Terraform.</li><li>Proficiency with common scripting languages such as Bash and Python.</li><li>United States citizenship and eligibility for a Secret clearance, since the work reaches export controlled Government environments. An active clearance helps. We sponsor candidates who are eligible but not currently cleared.</li></ul><p>You do not need every item on this list. If you have most of it and work well with other people, apply.</p>Preferred Qualifications<ul><li>Time at a Government supercomputing center, national laboratory, or university research computing center.</li><li>Work with STIG, SCAP, Tenable, eMASS, or RMF, or time inside FedRAMP or Impact Level boundaries.</li><li>Running GPU workloads on Kubernetes or OpenShift with Helm and operators, and using HPC containers such as Apptainer, Enroot, or Pyxis.</li><li>Running PBS Pro alongside Slurm, and monitoring with Prometheus and Grafana, including utilization and chargeback reporting.</li></ul> <p>Medical, vision, and dental coverage, a 401(k) with company match, short term disability, and generous paid vacation and sick time.</p>Equal employment opportunity<p>Parallel Works is an equal opportunity employer. We consider all qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, protected veteran status, or any other characteristic protected by law.</p>