Expert GPU Kernel Engineer

Expert GPU Kernel Engineer

Contract Type: Full-time, remote (worldwide)
Duration: Long-term
Monthly Rate: Up to $4500 USD

About the Role

We are looking for an Expert GPU Kernel Engineer to join a high-performing R&D team. The role involves developing and optimizing novel algorithms through low-level GPU kernels. You will work closely with algorithmic researchers to push performance boundaries and outperform state-of-the-art (SOTA) solutions.

Key Responsibilities

• Design and implement novel algorithms on GPUs, pushing hardware to its performance limits.
• Develop and apply advanced speedup and automation techniques.
• Build, maintain, and operate low-level profiling and benchmarking environments.
• Continuously monitor hardware innovations and leverage new GPU capabilities.
• Collaborate effectively with researchers and engineers to translate algorithmic concepts into highly optimized implementations.
   

Requirements

• Strong programming skills in C/C++ and CUDA.
• Proven experience with code profiling, debugging, and performance optimization.
• Excellent analytical and problem-solving skills.
• Ability to work independently and adapt quickly to new technologies.
   

Nice to Have

• Strong mathematical background.
• Deep understanding of AI and machine learning algorithms.
• MSc or PhD in Computer Science, Electrical Engineering, Mathematics, or a related field.