Senior AI HPC Systems Engineer

Placements24; Soweto; On-site; ZAR 600,000 - 900,000; Full time

Job summary

Placements24 is partnering with a cutting-edge AI research institution in Soweto, Gauteng, to hire an AI Systems Administrator. You will manage HPC clusters, GPUs, and AI workloads, ensuring reliability, scalability, and performance.

The role requires Linux administration expertise, HPC experience, and scripting skills to automate provisioning and operations. You'll collaborate with AI researchers and engineers to maintain a robust compute environment that accelerates research while staying

Qualifications

• Bachelor's degree in CS/IT or equivalent practical experience.
• Minimum 3 years in Linux admin with HPC focus.
• Strong server hardware, networking, and storage knowledge.
• Experience with Slurm or PBS cluster managers.
• Familiarity with AI/ML frameworks and GPU computing (CUDA).
• Scripting in Bash, Python, or Ansible.
• Excellent troubleshooting and teamwork.

Responsibilities

• Administer, maintain, and optimize HPC clusters, including compute nodes, storage systems, and networking.
• Install, configure, and manage AI/ML software frameworks and libraries (TensorFlow, PyTorch, CUDA).
• Monitor system performance, identify bottlenecks, and improve efficiency.
• Provide technical support for AI researchers and engineers using HPC.
• Manage user accounts, permissions, and resource allocation.
• Implement and maintain backup and disaster recovery strategies.
• Stay current with HPC and AI hardware advancements.
• Develop automation scripts for provisioning and configuration management.

Skills

Linux admin

HPC environments

AI/ML workloads

Slurm/PBS

CUDA GPUs

Bash/Python/Ansible

Networking & storage

Team collaboration

Education

Bachelor's in CS/IT

Tools

Slurm

PBS

CUDA

Job description

Placements24 is partnering with a cutting-edge AI research institution in Soweto, Gauteng, to hire an AI Systems Administrator. You will manage HPC clusters, GPUs, and AI workloads, ensuring reliability, scalability, and performance.

The role requires Linux administration expertise, HPC experience, and scripting skills to automate provisioning and operations. You'll collaborate with AI researchers and engineers to maintain a robust compute environment that accelerates research while staying