HPC Performance & Validation Engineer
Location: Dallas, TX
Overview
Our client is seeking an experienced
HPC Performance & Validation Engineer to help ensure large-scale GPU infrastructure is ready for production and consistently delivers the performance required for demanding AI, machine learning, and research workloads.
This role combines
GPU performance engineering, infrastructure benchmarking, automated validation, and observability across compute, storage, and networking. The engineer will establish testing standards, build repeatable validation frameworks, and investigate performance bottlenecks across a distributed HPC environment.
Working within the Architecture organization, this individual will partner with engineering, infrastructure, and research teams to turn benchmark results into practical improvements and guide future infrastructure decisions.
Key ResponsibilitiesGPU Validation & Performance Engineering
- Design and implement validation frameworks to verify GPU node readiness, utilization, and production performance.
- Establish repeatable methodologies for evaluating AI/ML workload performance across large-scale GPU clusters.
- Develop and execute industry-standard and workload-specific benchmarks across compute, storage, and networking.
- Investigate performance bottlenecks, identify root causes, and coordinate improvements with the appropriate engineering teams.
- Establish baseline performance metrics and continuous validation practices to measure reliability and efficiency as infrastructure evolves.
Automation & Observability
- Build scalable validation tools and micro-benchmarking frameworks using Python, Go, and Kubernetes.
- Integrate automated testing and benchmarking into CI/CD pipelines.
- Implement monitoring and dashboards to track cluster health, utilization, and performance using Prometheus, Grafana, OpenTelemetry, and ELK.
- Define an observability strategy that supports performance analysis, troubleshooting, and ongoing infrastructure validation.
Architecture & Cross-Functional Collaboration
- Translate benchmark findings into recommendations for infrastructure design, tuning, and capacity decisions.
- Document testing methodologies, hardware evaluations, and performance findings in technical reports.
- Partner with engineering, infrastructure, and research teams to align validation efforts with workload requirements and business priorities.
- Evaluate emerging hardware, tools, and architectures to inform long-term infrastructure planning.
Required Qualifications
- Experience profiling and tuning large-scale GPU clusters and accelerator-based infrastructure.
- Strong knowledge of NVIDIA ClusterKit, Nsight, NVIDIA validation tools, MLPerf, and DCGM, including GPU and DPU performance assessment.
- Experience benchmarking and optimizing network and storage performance across InfiniBand and RoCE environments using ClusterKit, iPerf, or comparable tools.
- Hands-on experience with Linux system benchmarking, including the Phoronix Test Suite or equivalent.
- Strong proficiency developing automation and micro-benchmarking tools using Python, Go, and Kubernetes in an Ubuntu Linux environment.
- Experience supporting HPC workloads across geographically distributed environments and using performance data to guide architectural decisions.
- Strong knowledge of OpenTelemetry, Prometheus, ELK, and Grafana, with experience defining observability practices for HPC infrastructure.
- Ability to evaluate emerging technologies and incorporate relevant findings into infrastructure strategy.
- Demonstrated ability to lead complex technical initiatives, communicate findings, and influence decisions across engineering and research teams.
Education
Bachelor’s degree in Computer Science, Engineering, or a related field, or equivalent professional experience.