HPC Systems Administrator

Numerical Algorithms Group•,
•Onsite

About The Position

The HPC Systems Administrator will support a major oil and gas client’s large GPU-based cluster for scientific computing, data analytics and AI. The role is dedicated to one client and covers GPU-accelerated clusters, high-speed networks, parallel storage and the automation that connects these systems. The administrator will work with scientists, AI engineers, software developers, infrastructure teams and technology suppliers, with a particular focus on making AI workloads perform well on shared HPC infrastructure. The position offers hands-on responsibility for troubleshooting issues such as stalled training runs and performance degradation, and contributing to the platform’s development. The role involves deploying, operating, securing and improving HPC environments. It includes Linux and GPU systems administration; scheduler, network, storage, monitoring and automation support; deployment of reproducible AI environments; performance investigation; and operational support for business-critical infrastructure. The position is on-site in Houston, Texas, and is full-time. Relocation is not offered.

Requirements

  • Bachelor’s degree in computer science, computer engineering or a related engineering discipline.
  • At least five years of experience supporting HPC, research computing, AI infrastructure or large-scale Linux environments.
  • Practical Linux system administration experience, including installation, configuration, patching, security and troubleshooting.
  • Experience with a workload scheduler or resource-management platform.
  • Familiarity with high-speed networks, parallel or scale-out storage, and distributed applications.
  • Scripting and automation experience using Python and Bash.
  • Experience with source control and configuration-management or CI/CD practices.
  • Experience supporting scientists, engineers, developers or other technical users.
  • Willingness to participate in maintenance, incident response and operational support for business-critical infrastructure.
  • Working knowledge of HPC infrastructure, including compute nodes, GPUs, memory, interconnects, workload schedulers, parallel filesystems and cluster-management services.
  • Ability to deploy and operate reproducible environments for AI training, fine-tuning, evaluation and inference on shared GPU infrastructure.
  • Experience with software-environment tools such as Python, Conda, Spack, environment modules, Jupyter and GPU software libraries.
  • Experience running scientific and AI software through HPC-compatible container technologies such as Apptainer/Singularity.
  • Familiarity with MPI or multi-GPU and multi-node AI workloads.
  • Ability to troubleshoot bottlenecks across applications, compute, GPU memory, networks, storage and schedulers.
  • Ability to automate routine administration and support monitoring, logging and alerting.
  • Understanding of access control, patching, vulnerability remediation, secrets, software provenance and data protection.
  • Ability to document technical work and communicate with infrastructure specialists and scientific users.

Nice To Haves

  • Experience supporting GPU-accelerated systems using NVIDIA CUDA or AMD ROCm.
  • Familiarity with AI frameworks such as PyTorch, TensorFlow or JAX.
  • Experience running multi-GPU or multi-node AI workloads using NCCL, RCCL or similar technologies.
  • Experience integrating GPU workloads with Slurm or another scheduler.
  • Familiarity with AI inference platforms such as NVIDIA Triton Inference Server or vLLM, and AI/LLM gateways such as Bifrost or LiteLLM.
  • Experience with MLOps capabilities such as experiment tracking, model registries or artifact management.
  • Experience supporting secure multi-user Jupyter or Open OnDemand environments.
  • Experience with InfiniBand, MPI performance, parallel filesystems or data-intensive scientific workflows.
  • Familiarity with retrieval-augmented generation, scientific AI or enterprise AI services.

Responsibilities

  • Deploy, operate, secure and continually improve HPC environments supporting scientific computing, data analytics and AI.
  • Administer Linux clusters, GPU systems, high-speed networks, parallel storage, scheduling, monitoring and automation.
  • Work with scientists, AI engineers, developers, infrastructure teams and suppliers to investigate and resolve system and application issues.
  • Help design and implement reliable, scalable computing services that are straightforward to support.
  • Deploy AI workloads on shared HPC infrastructure, including GPU resource management, containerized software environments, distributed training and inference, and secure user access.
  • Investigate and tune performance across compute, memory, network and storage.
  • Contribute to platform maintenance, upgrades, automation, documentation and technology evaluations.
  • Provide operational support, including maintenance windows and incident response for business-critical infrastructure.
  • Support users and help them get more from the platform.
  • Improve monitoring and alerting, and automate routine administration tasks.

Benefits

  • 401(k) plan with company match up to 5%.
  • Health insurance.
  • Dental insurance.
  • Life insurance.
  • Short-term and long-term disability insurance.
  • Employee-paid vision insurance is available.
  • 9 public holidays.
  • 25 vacation days, of which 5 are fixed over Christmas and New Year and 20 can be scheduled by the employee.
  • Paid sick days.
  • Maternity and paternity leave.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service