Remote CUDA Kernel Optimization Engineer

Mercor; Germany (OH); On-site; USD 110,208 - 165,312; Part time

Job summary

Mercor is looking for a CUDA Engineering Expert to join their team and work remotely. This role will focus on analyzing and optimizing GPU kernels to enhance performance and efficiency.

The ideal candidate should possess strong experience in C++17, Python, and GPU programming. Key responsibilities include documenting optimization decisions and improving kernel implementations using profiler metrics. Candidates must be available for at least 20 hours a week. The position offers competitive compensation between $80 to $120 per hour.

Qualifications

• Available to work at least 20 hours per week.
• Strong understanding of GPU profiler performance metrics for optimization.
• At least 1 year of relevant research experience with GPUs.

Responsibilities

• Analyze and optimize GPU kernels for performance and efficiency.
• Use profiler metrics to guide improvements on kernel implementations.
• Document optimization decisions clearly.

Skills

CUDA proficiency

C++17 fluent

Python knowledge

GPU programming model experience

Tools

Git

NSight Compute

Job description

Mercor is looking for a CUDA Engineering Expert to join their team and work remotely. This role will focus on analyzing and optimizing GPU kernels to enhance performance and efficiency.

The ideal candidate should possess strong experience in C++17, Python, and GPU programming. Key responsibilities include documenting optimization decisions and improving kernel implementations using profiler metrics. Candidates must be available for at least 20 hours a week. The position offers competitive compensation between $80 to $120 per hour.