HPC/AI Network Engineer — Equity Eligible

Supermicro; San Jose (CA); On-site; USD 120,000 - 140,000; Full time

Job summary

Supermicro in San Jose, CA seeks a Network Engineer to assist roll out and maintain business-critical applications and services. You will work with a senior engineer to resolve escalated issues and implement complex projects.

Candidate will test hardware and software, document procedures, and support HPC/AI workloads, leveraging Linux, Docker, and Kubernetes, with a focus on reliability and performance.

Qualifications

• BS/MS in Electrical Engineering, Computer Engineering or Computer Science.
• 1+ years of work-related experience in Deep Learning and Machine Learning.
• Familiar with Linux networking debugging/testing or relevant experience.
• Familiar with data center, enterprise, or telecommunication working on routing and switching networking technologies.
• Knowledge with DevOps or in cloud environments, including Docker/Containers and Kubernetes.
• Hands-on experience with workload/scheduler Managers (Slurm) for rack/cluster.
• Familiar with MLPerf Training/Inference benchmark, LLM, HPL-AI or RCCL/NCCL.
• Programming experience with Windows and Linux shell scripting.
• Strong sense of teamwork and good communication skills.

Responsibilities

• Execute system-level rack tests on latest NVidia and AMD GPUs, ARM-based, Intel Xeon, and AMD EPYC processors, encompassing functionality, compatibility, performance, stress, and reliability testing, leveraging proprietary in-house tools.
• Familiar with HPC/AI applications and benchmarks, address customer support issues, demonstrating innovative problem-solving skills and building robust processes and procedures for HPC/AI solutions.
• May work on conduct proof of concept design and testing, providing optimized benchmarks for HPC/AI applications in a timely manner. Fine-tune BIOS settings, optimize OS/network configurations, and develop diverse simulation configurations to enhance efficiency across various workloads.
• Deliver on-site deployment services, ensuring customer acceptance verification and providing post-level 1&2 support. Create and maintain technical documentation, including technical notes, blogs, and diagrams, to facilitate knowledge dissemination.
• Identify and document hardware and software quality issues and collaborate with Product Management and other Engineering teams to integrate customer feedback into future product enhancements.
• Proactively engage in HPC roadmap development, planning software and hardware upgrades to sustain exceptional HPC infrastructure performance.
• Document and analyze test plans, reports, logs, and actively contribute to the development of test utilities and automation scripts to streamline testing processes.

Skills

Deep Learning

Linux Networking

Data Center Networking

DevOps / Cloud

Slurm workload Manager

Scripting

Teamwork/Communication

Education

BS/MS in Electrical/Computer Engineering or CS

Tools

Docker

Kubernetes

Linux

OpenStack/OpenShift

CUDA/oneAPI/ROCm

Job description

Supermicro in San Jose, CA seeks a Network Engineer to assist roll out and maintain business-critical applications and services. You will work with a senior engineer to resolve escalated issues and implement complex projects.

Candidate will test hardware and software, document procedures, and support HPC/AI workloads, leveraging Linux, Docker, and Kubernetes, with a focus on reliability and performance.