GPU Software Engineer (HPC / Deep Learning Optimization)

Luxoft; Town of Poland (NY); On-site; USD 120,000 - 180,000; Full time

Job summary

Luxoft is seeking a Software Engineer focused on GPU computing, HPC workloads, and Deep Learning inference optimization on Windows.

You will develop and optimize GPU compute kernels in C++ using CUDA or similar, profile workloads, and collaborate with engineers to improve performance and stability.

Experience with CUDA/OpenCL and profiling tools is highly valued; working knowledge of NVIDIA/AMD toolchains and Windows GPU development is a plus.

Qualifications

• Hands-on experience writing and optimizing GPU compute kernels in CUDA, OpenCL, SYCL, or equivalent.
• Ability to profile and performance tune GPU code (memory, compute, latency) as part of that work.

Responsibilities

• Develop, optimize, and maintain GPU compute kernels using C++ and a GPU programming framework.
• Profile GPU workloads and tune memory, compute, and latency to improve performance and efficiency.
• Analyze performance bottlenecks and apply targeted optimizations.
• Debug and resolve performance and stability issues.
• Apply kernel optimization to HPC or Deep Learning inference pipelines where needed.
• Collaborate with engineers, QA, and stakeholders.
• Follow coding standards and contribute to technical documentation.

Skills

GPU kernel programming

CUDA/OpenCL/HIP/SYCL

Performance profiling

Windows development

C++ proficiency

Tools

Nsight

Radeon GPU Profiler

PIX

TensorRT

ONNX Runtime

PyTorch

Job description

We are looking for a Software Engineer focused on GPU computing, HPC workloads, and Deep Learning inference optimization on Windows platform. The project is aimed at improving performance and efficiency of GPU-based workloads, including compute kernels and inference pipelines. The role is not limited to graphics APIs and is suitable for candidates with strong experience in CUDA, OpenCL, or similar technologies, as well as shader-based optimization.

Responsibilities

• Develop, optimize, and maintain GPU compute kernels using C++ and a GPU programming framework (CUDA, HIP, OpenCL, SYCL, DirectCompute / HLSL compute shaders, Metal compute, or equivalent).
• Profile GPU workloads and tune memory, compute, and latency to improve performance and efficiency.
• Analyze performance bottlenecks and apply targeted optimizations.
• Debug and resolve performance and stability issues.
• Apply kernel optimization to HPC or Deep Learning inference pipelines where needed.
• Collaborate with engineers, QA, and stakeholders.
• Follow coding standards and contribute to technical documentation.

Mandatory Skills Description

• Hands on experience writing and optimizing GPU compute kernels in at least one framework: CUDA, HIP, OpenCL, SYCL, DirectCompute / HLSL compute shaders, Metal compute, or equivalent.
• Ability to profile and performance tune GPU code (memory, compute, latency) as part of that work.

Nice-to-Have Skills Description

• Strong knowledge of C++
• Experience with HPC or Deep Learning inference optimization
• Experience with profiling tools (Nsight, Radeon GPU Profiler, PIX, etc.)
• Experience with Deep Learning frameworks (TensorRT, ONNX Runtime, PyTorch, DirectML, etc.)
• Understanding of graphics pipelines and rendering basics
• Experience with a graphics API (DirectX, Vulkan, Metal, etc.)
• Experience with Windows platform GPU development
• Experience with CI/CD, version control, or automated testing