Energy HPC CUDA Performance Engineer

NVIDIA; United Arab Emirates; On-site; AED 420,000 - 660,000; Full time

Job summary

NVIDIA is seeking a Compute Developer Technology (DevTech) expert to accelerate energy simulations and AI workflows. You will optimize CUDA performance on multi-GPU and multi-node systems and work with customer teams to deliver measurable speedups.

The role requires strong C/C++ and Python, CUDA programming, and experience with Nsight tools for profiling. You will engage with internal teams to influence roadmaps and build reusable libraries for future NVIDIA products.

Qualifications

• BS/MS or equivalent in CS/CE/EE/Physics/Applied Math or related field.
• Strong programming skills in C/C++ and Python on Linux.
• Hands-on CUDA programming and GPU performance optimization experience.
• Experience with profiling using NVIDIA Nsight Systems / Nsight Compute.

Responsibilities

• Profile, analyze, and optimize GPU-accelerated applications focusing on CUDA kernels and memory movement.
• Drive performance improvements across the stack, including MPI/NCCL multi-GPU/multi-node scaling.
• Build reproducible benchmarks, performance reports, and tuning recommendations.
• Develop reference implementations and patches to enable performance and portability.
• Support customer engagements from POC to production and advise on deployment best practices.

Skills

C/C++

Python

CUDA programming

Profiling

MPI/NCCL

Linux

English

HPC optimization

Nsight tools

Education

BS/MS in CS/CE/EE/Physics/Applied Math

Tools

Nsight Systems

Nsight Compute

Job description

NVIDIA is seeking a Compute Developer Technology (DevTech) expert to accelerate energy simulations and AI workflows. You will optimize CUDA performance on multi-GPU and multi-node systems and work with customer teams to deliver measurable speedups.

The role requires strong C/C++ and Python, CUDA programming, and experience with Nsight tools for profiling. You will engage with internal teams to influence roadmaps and build reusable libraries for future NVIDIA products.