HPC Performance & Validation Engineer (Hybrid, Dallas)

Gosnaphop; Dallas (TX); Hybrid; USD 180,000 - 260,000; Full time

Benefits offered by this job

100% paid medical, dental, vision

401(k)

25 days PTO

HSA contribution

Lunch on office days

Gym membership

Job summary

Gosnaphop is seeking an HPC Performance and Validation Engineer in Dallas, TX to measure, validate, and improve distributed HPC system performance. You will build repeatable GPU readiness tests for production workloads, investigate bottlenecks, and guide infrastructure decisions with data.

The role uses Python/Go, Kubernetes, and CI/CD to create validation workflows, with hybrid work and relocation options. A strong focus on performance, profiling, and observability is required.

Qualifications

• Experience benchmarking, profiling, and tuning large-scale GPU clusters and HPC workloads.
• Familiar with accelerator performance tools and GPU debugging/test utilities.
• Proficiency building automated validation tools with Python or Go on Linux; Kubernetes experience is needed.
• Knowledge of observability tools (Prometheus, Grafana, OpenTelemetry, ELK) is important.

Responsibilities

• Build automated tests that verify GPU node readiness, health, and utilization across a large HPC environment.
• Benchmark compute, storage, and network performance with established tests.
• Profile AI/research workloads; identify bottlenecks and drive improvements with teams.
• Develop validation tools and CI/CD workflows using Python, Go, Kubernetes.
• Establish monitoring and reporting to show cluster health and performance trends.
• Document benchmark findings to guide architecture and capacity decisions.
• Lead technical investigations and collaborate with infrastructure and research teams.

Skills

GPU benchmarking

Performance profiling

Technical leadership

Python programming

Go programming

Education

Bachelor's degree in a relevant field

Tools

NVIDIA Nsight

DCGM

ClusterKit

MLPerf

Prometheus

Grafana

OpenTelemetry

ELK stack

Kubernetes

Python

Go

Job description

Gosnaphop is seeking an HPC Performance and Validation Engineer in Dallas, TX to measure, validate, and improve distributed HPC system performance. You will build repeatable GPU readiness tests for production workloads, investigate bottlenecks, and guide infrastructure decisions with data.

The role uses Python/Go, Kubernetes, and CI/CD to create validation workflows, with hybrid work and relocation options. A strong focus on performance, profiling, and observability is required.