Technical articles tagged CUDA, covering implementation details and research notes by Hwan Heo.
From sparse quadratic solvers in quad remeshing to deep learning optimizers
Why naive GEMM becomes memory-bound, how tiled reuse changes arithmetic intensity, and where Tensor Cores, schedulers, and libraries fit
Why generic compiler paths underperform on dynamic sparse 3D generative models—and how null-context attention removal and model-aware CUDA kernels cut inference