Audio upscaling detector and command line spectrum visualisation tool
A hyper-accelerated C++ AVX2 SIMD raycasting engine & closed-loop 5-layer Deep RL architecture for low-power edge robotics (<15W CPU-only). Outperforms Yamauchi frontier exploration with 84.5% map coverage & 93.5% collision reduction.
IAA03_fast_math is a single-header math kernel(Atan2 only for now) designed to eliminate the "Trigonometry Tax" in high-throughput systems (Physics Engines, Audio DSP, and ML Pre-processing). Branchless ,ILP and SIMD (AVX2/SSE4.1), it achieves up to a ~186x per-element throughput speedup over std::atan2 while being IEEE 754 compliant
A resurrection of a very old project, brought back to life to save my brain.
Optimized Firefox ESR for Windows built from source with PGO/LTO for ZEN5 and AVX2 CPUs
High-performance matrix multiplication engine in C/C++, leveraging Cache Tiling, ILP Loop Unrolling, 256-bit SIMD (AVX2/FMA), and 14-core OpenMP to achieve a 3,176.44× speedup.
Cache-conscious sorted bucket data structure with pressure-flow rebalancing, AVX2 SIMD, and OLC concurrency
A optimized Game of Life generation.
Fast-Fourier Transform C++26 implementation
Safety-hardened GEMM (matrix multiply) implementation achieving 169.8 GFLOPS on Intel i9-14900. Built for embedded systems and safety-critical applications where reliability matters as much as speed. 162× faster than naive, zero UB, fully validated.
Implementing the ChaCha20-Poly1305 authenticated encryption scheme in C++, optimizing it with AVX2 intrinsics against OpenSSL as the baseline, and extending it to the prime field 2^150 − 3.
To associate your repository with the avx2-optimized topic, visit your repo's landing page and select "manage topics."