CUDA Kernel Optimization Engineer

Optimize GPU kernels for a contract-based AI research project, using profiler-guided analysis to improve performance, efficiency, and hardware utilization across modern GPU environments.

This per-task opportunity is suited to GPU programming specialists who enjoy diagnosing bottlenecks and improving low-level kernel performance. You will evaluate, optimize, and reason about GPU kernels without needing extensive prior context for every underlying algorithm.

• Analyze and optimize GPU kernels for performance, efficiency, and hardware utilization.
• Use metrics such as L2 cache hit rate, L2 throughput, occupancy, and related profiler signals to guide improvements.
• Review kernel implementations, identify bottlenecks, and recommend performance changes.
• Write, modify, and reason about C++17, Python, and GPU programming code.
• Apply CUDA, HIP, shader programming, or related kernel-programming expertise to improve results.
• Clearly document optimization decisions, including when specific profiler metrics are or are not useful.

• Fluency in core C++ features through C++17.
• Working knowledge of Python and Git.
• Fluency in at least one GPU programming model, including CUDA, HIP, Slang, HLSL, GLSL, or a related kernel-programming technology.
• At least 1 year of professional or graduate-level research experience working with GPUs.
• Strong understanding of GPU profiler performance metrics and their use in kernel optimization.
• Ability to optimize kernels without deep prior knowledge of every algorithm.
• Experience with CUDA, HIP, CUDA C++ Core Libraries, inline PTX assembly, or tensor core-level optimization is a plus.
• Experience optimizing kernels for NVIDIA Blackwell hardware is a plus.
• Familiarity with NSight Compute is a plus.
• Prior experience with GPU hardware organizations such as NVIDIA, AMD, or Qualcomm is a plus.
• Open-source contributions related to GPU kernel optimization are a plus.

• Remote, per-task contract engagement.
• Availability of at least 20 hours per week is required.

$500 per task.

• Submit a resume or relevant technical background.
• Qualified applicants may be asked to complete a brief technical assessment or provide additional information.