GPU Programmer

I'm looking for a seasoned professional to design, develop, and optimize GPU-based algorithms specifically for enhancing the accuracy of AI computations. The role will heavily involve working with matrix factorization within numerical linear algebra, identifying and solving performance bottlenecks predominantly related to kernel execution time. Key Responsibilities: - Create and implement high-performance GPU algorithms aimed at improving AI calculation precision. - Collaborate with mathematicians and AI researchers to innovate linear algebra solutions. - Diagnose and rectify performance issues, with a focus on kernel execution time constraints. - Contribute to the architectural development of our main acceleration platform. Ideal Skills and Experience: - Extensive background in CUDA and C/C++ is essential. - Proven expertise in parallel programming and performance optimization. - Prior experience with numerical linear algebra, specifically matrix factorization. - Exceptional problem-solving abilities and meticulous attention to detail. - Self-sufficient with a capacity to work with minimal oversight. Technologies Used: - C++ - CUDA - GPU Programming # Note: This is a contractor role, employee benefits are not included. To apply, please send your CV detailing relevant experience and completed projects.