AI and HPC Systems Performance Engineer

Overview

AI and HPC Systems Performance Engineer focused on tuning GPU server performance for AI training and inference workloads on Linux-based systems.

What you'll do

• Install, configure, and optimize AI infrastructure components including GPU servers, storage, networking, and AI software stacks.
• Capture, analyze, and interpret telemetry, logs, traces, and profiling data to characterize workload execution and find bottlenecks.
• Design, execute, and analyze performance benchmarks for AI/ML workloads including LLMs and RAG pipelines.
• Optimize performance across multi-GPU and distributed AI environments using interconnect, storage, and networking technologies.
• Develop automation scripts, deployment frameworks, and Infrastructure-as-Code solutions for AI platform provisioning.
• Collaborate with customers, partners, and internal engineering teams to troubleshoot and optimize AI solutions on HPE platforms.

What you'll need

• Typically 8+ years of experience.
• Strong Linux system administration and command-line experience across enterprise Linux distributions.
• Experience with modern AI/ML frameworks and ecosystems including PyTorch, JAX, and Hugging Face Transformers.
• Experience with AI training/inference, benchmarking, performance characterization, and optimization.
• Experience with high-performance networking technologies including InfiniBand and RDMA (incl. RoCE) and networking solutions.
• Proficiency in one or more programming/scripting languages such as Python, Bash, Go, C++, or similar.
• Experience with containerized/orchestrated environments including Docker and Kubernetes.

Details

• Hybrid role with an average requirement to work 2 days per week from an HPE office.