A personal blog series covering CUDA GPU programming, from optimizing single-precision matrix multiplication to low-precision arithmetic in ML and profiling kernels. Readers get tutorial-style explanations of GPU performance concepts, with code mostly pointed to external GitHub and Medium links.