Working mostly on CUDA Projects day-to-day since last 5 years, so have good understanding of architecture and algorithmic level cuda program optimization.
Having extensively worked on all NVIDIA GPU architectures till Pascal , know all the possible cuda features and architecture level optimization.
I am pretty sure I can get you optimal performance out of your application for NVIDIA GPUs.
Though I am new in freelancer provided an opportunity I can prove my value for the project, I am very much confident of delivering optimal performance efficient code.