https://veitner.bearblog.dev/ 2026-03-11 https://veitner.bearblog.dev/simple-math-to-speed-up-gdn-prefill/ 2026-03-14 https://veitner.bearblog.dev/chunkwise-gated-delta-rule/ 2026-03-10 https://veitner.bearblog.dev/gated-delta-net-decoding/ 2026-02-15 https://veitner.bearblog.dev/grouped-blockscaled-gemm-kernel/ 2026-02-07 https://veitner.bearblog.dev/grouped-blockscaled-gemm-host-code/ 2026-01-24 https://veitner.bearblog.dev/grouped-block-scaled-gemm-intro/ 2026-01-22 https://veitner.bearblog.dev/warp-specialisation-in-cutedsl/ 2026-01-07 https://veitner.bearblog.dev/2-cta-gemm-on-b200/ 2026-01-04 https://veitner.bearblog.dev/blackwell-pipelining-with-cutedsl/ 2025-12-23 https://veitner.bearblog.dev/b200-blockscaled-gemm-the-setup/ 2025-12-06 https://veitner.bearblog.dev/scale-tensor-construction-in-cutedsl/ 2025-12-03 https://veitner.bearblog.dev/demystifying-numeric-conversions-in-cutedsl/ 2025-11-23 https://veitner.bearblog.dev/nvfp4-gemv-improved/ 2025-11-16 https://veitner.bearblog.dev/nvfp4-gemv/ 2025-11-13 https://veitner.bearblog.dev/bit-counting-and-geometric-series/ 2025-11-06 https://veitner.bearblog.dev/simple-reduction-in-cutedsl/ 2025-10-27 https://veitner.bearblog.dev/mma-atoms-in-cute/ 2025-12-12 https://veitner.bearblog.dev/lru-and-lfu-in-c/ 2025-10-11 https://veitner.bearblog.dev/mutual-refinement-and-composition/ 2025-10-03 https://veitner.bearblog.dev/applied-introduction-to-categorical-treatment-of-cute/ 2025-09-27 https://veitner.bearblog.dev/layout-gymnastics/ 2025-09-23 https://veitner.bearblog.dev/swizzles-and-their-usage-in-cutedsl-kernels/ 2025-09-20 https://veitner.bearblog.dev/cute-partitions/ 2025-09-14 https://veitner.bearblog.dev/tensors-slicing-in-cute/ 2025-09-09 https://veitner.bearblog.dev/understanding-cute-swizzling-the-math-behind-32b-64b-and-128b-patterns/ 2026-01-16 https://veitner.bearblog.dev/gpu-l2-cache-persistence/ 2025-09-05 https://veitner.bearblog.dev/cuda-streams/ 2025-09-01 https://veitner.bearblog.dev/pingpong-in-the-cutedsl-with-quack/ 2025-08-23 https://veitner.bearblog.dev/bit-hacking-in-c/ 2025-08-17 https://veitner.bearblog.dev/intuition-behind-hierarchical-layouts/ 2025-08-14 https://veitner.bearblog.dev/persistent-float8-dense-gemm-on-hopper/ 2025-08-09 https://veitner.bearblog.dev/epilogue-h100-cutedsl/ 2025-12-12 https://veitner.bearblog.dev/let-the-compiler-do-the-work-in-cutedsl/ 2025-07-30 https://veitner.bearblog.dev/persistent-gemm-in-cutedsl-on-hopper/ 2025-07-25 https://veitner.bearblog.dev/consumer-producer-pattern-on-h100-in-cutedsl/ 2025-07-20 https://veitner.bearblog.dev/backprob-through-layernorm/ 2025-07-13 https://veitner.bearblog.dev/backprop-through-rmsnorm/ 2025-07-13 https://veitner.bearblog.dev/outperform-compiled-pytorch-code-using-quack/ 2025-07-12 https://veitner.bearblog.dev/cutedsl-on-hopper-pipelining/ 2025-07-05 https://veitner.bearblog.dev/cutedsl-on-hopper-wgmma-and-tma-intro/ 2025-12-29 https://veitner.bearblog.dev/thread-value-layouts-in-cute/ 2025-06-28 https://veitner.bearblog.dev/sgemm-in-cutedsl/ 2025-06-26 https://veitner.bearblog.dev/an-applied-introduction-to-cutedsl/ 2025-06-24 https://veitner.bearblog.dev/calculating-the-fibonacci-numbers-on-gpu/ 2025-06-21 https://veitner.bearblog.dev/an-introduction-to-thrust/ 2025-06-17 https://veitner.bearblog.dev/infinite-binary-string/ 2025-06-09 https://veitner.bearblog.dev/highly-efficient-matrix-transpose-in-mojo/ 2025-06-06 https://veitner.bearblog.dev/the-bijection-between-natural-numbers-and-binary-strings/ 2025-06-05 https://veitner.bearblog.dev/use-tma-without-cuda/ 2025-06-04 https://veitner.bearblog.dev/use-ptx-instructions-in-mojo/ 2025-05-29 https://veitner.bearblog.dev/very-fast-vector-sum-without-cuda/ 2025-05-25 https://veitner.bearblog.dev/short-introduction-to-the-mojo-programming-language/ 2025-05-22 https://veitner.bearblog.dev/bridging-math-and-code-cute-layout-algebra-in-cutedsl/ 2025-05-18 https://veitner.bearblog.dev/load-and-store-matrices-efficently-with-ptx-instructions/ 2025-12-12 https://veitner.bearblog.dev/how-to-use-reasoning-models-with-sglang/ 2025-05-11 https://veitner.bearblog.dev/making-matrix-transpose-really-fast-on-hopper-gpus/ 2025-05-07 https://veitner.bearblog.dev/tma-introduction/ 2025-04-27 https://veitner.bearblog.dev/analyze-cuda-programs-by-looking-at-gpu-assembly/ 2025-04-21 https://veitner.bearblog.dev/making-rmsnorm-really-fast/ 2025-04-18 https://veitner.bearblog.dev/making-prefix-sum-really-fast/ 2025-04-13 https://veitner.bearblog.dev/making-vector-sum-really-fast/ 2025-04-09 https://veitner.bearblog.dev/predication-in-cutlass/ 2025-03-31 https://veitner.bearblog.dev/indexing-in-cuda/ 2025-03-23