We are seeking a highly skilled HPC/GPU Operations Engineer to manage, optimize, and maintain high-performance computing (HPC) infrastructure, with a focus on GPU-accelerated workloads. The ideal candidate will be responsible for ensuring the reliability, efficiency, and scalability of HPC systems used for scientific computing, AI/ML, and data-intensive applications. With 6-10 years of experience
Career Level – IC3
HPC & GPU System Management
Administer and maintain HPC clusters, GPU nodes, and high-speed interconnects.
Deploy and configure GPU-accelerated workloads for AI/ML, scientific computing, and simulations.
Monitor system performance, troubleshoot issues, and optimize resource utilization.
Software & Middleware Support
Install, configure, and maintain HPC-related software, libraries, and tools (CUDA, OpenMP, MPI, etc.).
Support containerized workflows using Docker, Singularity, or similar technologies.
Ensure compatibility of software stacks with GPU architectures (NVIDIA, AMD, Intel).
Performance Optimization & Monitoring
Tune GPU and CPU performance for specific workloads, including benchmarking and profiling.
Utilize monitoring tools (e.g., Prometheus, Grafana, Slurm, Ganglia) to track system health and efficiency.
Optimize scheduling and resource allocation in workload managers (Slurm, PBS, LSF, etc.).
Security & Compliance
Ensure system security and access control for HPC resources.
Apply software patches, firmware updates, and security best practices.
Assist in regulatory compliance for HPC environments.
User Support & Documentation
Provide support to researchers, data scientists, and engineers using HPC resources.
Develop and maintain documentation on best practices, troubleshooting, and system usage.
Conduct training sessions or workshops on HPC/GPU computing.
Required Qualifications
Technical Skills
Experience managing HPC clusters and GPU-based computing environments.
Proficiency in Linux system administration, scripting (Bash, Python), and automation (Ansible, Terraform).
Knowledge of parallel computing, GPU programming (CUDA, OpenCL), and HPC frameworks.
Familiarity with networking (Infiniband, RDMA), storage (Lustre, GPFS, NFS), and virtualization.
An easy way to apply for this job. Use the following social media.
An easy way to apply for this job. Use the following social media.
