HPC Performance & Validation Architect

NorthMark Compute & Cloud; Dallas (TX); On-site; USD 110,000 - 170,000; Full time

Job summary

NorthMark Compute & Cloud is seeking an HPC Validation and Performance Engineer to own validation and optimization of HPC CPU/GPU farms. You will develop a validation and baselining framework for AI/ML workloads, drive performance benchmarking, and implement advanced tooling to scale our datacenter footprint.

You will lead cross-functional efforts, build automation using Python/Go/Kubernetes, and implement modern monitoring with Prometheus and Grafana to deliver data-driven insights for

Qualifications

• Bachelor's degree or equivalent experience required.
• Experience profiling and tuning with large GPU clusters.
• Deep knowledge of NVIDIA ClusterKit, Nsight, MLPerf and DCGM.
• Networking & storage perf; InfiniBand/RoCe profiling and optimization.
• Automation tooling and micro-benchmarking using Python and Go; Kubernetes in Ubuntu.
• Experience with HPC workloads across distributed global locations.

Responsibilities

• Architect and implement a validation framework for GPU nodes in a large HPC environment.
• Define methods to continually assess performance and optimize AI/ML workloads.
• Develop and execute comprehensive performance tests using benchmarks for HPC compute, storage and networking.
• Contribute to research reports describing benchmarking findings and HW performance.
• Lead bottleneck debugging and resolution efforts in system performance.
• Build scalable tools for automated validation and testing using Python, Go, Kubernetes and CI/CD.
• Implement monitoring with Prometheus, Grafana, OTEL and related technologies for real-time health.
• Define observability strategy for HPC validation and performance monitoring.
• Stay informed on industry trends to ensure long-term strategic alignment.
• Collaborate cross-functionally with engineering, infra and research teams.

Skills

Accelerator performance

Profiling & tuning

Automation tooling

Leadership

Cross-functional collaboration

Education

Bachelor's degree or equivalent experience

Tools

NVIDIA ClusterKit

Nsight

MLPerf

DCGM

Phoronix Test Suite

Python

Go

Kubernetes

iPerf

Prometheus

Grafana

Job description

NorthMark Compute & Cloud is seeking an HPC Validation and Performance Engineer to own validation and optimization of HPC CPU/GPU farms. You will develop a validation and baselining framework for AI/ML workloads, drive performance benchmarking, and implement advanced tooling to scale our datacenter footprint.

You will lead cross-functional efforts, build automation using Python/Go/Kubernetes, and implement modern monitoring with Prometheus and Grafana to deliver data-driven insights for