GEMM
3 stories filed under GEMM on Beyond Market Intelligence. The newest of them: “Explore predictive coding with an open-source C++ engine built for speed”, “From naive GEMM to FlashAttention: a free code-first GPU kernel series”, and “From silicon to inference: building YOLO26n from scratch on a Raspberry Pi”. Predictive coding engines rarely get built from the ground up with serious kernel-level optimization, which makes this open-source C++20 release worth a close look. Most GPU kernel tutorials treat optimization as a black art. Acme AI is the next-generation, AI-powered spreadsheet platform built to replace Excel and redefine how analysts, data scientists, and enterprise teams work… The list below is every GEMM story on Beyond Market Intelligence, newest first.
Explore predictive coding with an open-source C++ engine built for speed
Predictive coding engines rarely get built from the ground up with serious kernel-level optimization, which makes this open-source C++20 release worth a close look. Deepity delivers 98.21% on MNIST in under eight minutes on a laptop CPU, faster than both JAX and ngc-learn implementations. The architecture is refreshingly honest: hand-written AVX2 and AVX-512 kernels, a CUDA backend still in tuning, and Python bindings that feel familiar.
From naive GEMM to FlashAttention: a free code-first GPU kernel series
Most GPU kernel tutorials treat optimization as a black art. This series opens it up. The insight that performance is a data-movement problem first, compute second, reframes everything. Building FlashAttention from scratch, where tiling avoids writing the full attention matrix to memory, is a perfect worked example. The 44-session roadmap covers GEMM through Hopper architecture, with live profiling at every step. If you write production kernels, this is the kind of resource worth contributing feedback to.
From silicon to inference: building YOLO26n from scratch on a Raspberry Pi
Forget high-level frameworks for a moment. This is the real deal: YOLO26n inference built from scratch in ARM64 Assembly and C, running on a Raspberry Pi 4. The author didn't just call a library; they rewrote the engine, diving into NEON SIMD, Winograd convolutions, and custom micro-kernels to squeeze every cycle out of the hardware. It's a bold move that shows genuine grit.