Bruce-Lee-LY

LLM Infer, AI Infra, CUDA

GitHub 18 repos 233 followers

Explained projects

memory_pool: A C++ Memory Pool That Chooses Clarity Over Cleverness
memory_pool: A C++ Memory Pool That Chooses Clarity Over Cleverness
A tiered power-of-two allocator with a simple API, thread-safe blocks, and a surprisingly linear free path that makes its trade-offs easy to see.
7 min read · May 2, 2026

Yet to be explained

cuda_hgemm
Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruction.
Cuda569 stars
Explain
cuda_hook
Hooked CUDA-related dynamic libraries by using automated code generation tools.
C173 stars
Explain
cuda_hgemv
Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.
Cuda76 stars
Explain
decoding_attention
Decoding Attention is specially optimized for MHA, MQA, GQA and MLA using CUDA core for the decoding stage of LLM inference.
C++48 stars
Explain
flash_attention_inference
Performance of the C++ interface of flash attention and flash attention v2 in large language model (LLM) inference scenarios.
C++46 stars
Explain
cuda_auto_tune
NCU-driven iterative optimization workflow for CUDA/CUTLASS/Triton/CuTe DSL kernels.
Python25 stars
Explain
cutlass_gemm
Multiple GEMM operators are constructed with cutlass to support LLM inference.
C++20 stars
Explain
matrix_multiply
Several common methods of matrix multiplication are implemented on CPU and Nvidia GPU using C++11 and CUDA.
C++14 stars
Explain
cuda_back2back_hgemm
Use tensor core to calculate back-to-back HGEMM (half-precision general matrix multiplication) with MMA PTX instruction.
Cuda13 stars
Explain
DeepGEMMPerTensor
DeepGEMMPerTensor: clean and efficient FP8 GEMM per tensor kernels without scales
Python5 stars
Explain
thread_pool
Thread pool is implemented to process task queue with C++11.
C++3 stars
Explain
deep_learning
Implemented the training and inference of several common deep learning model algorithms with tensorflow and pytorch.
Python1 stars
Explain
algorithm_design
Use several algorithm design methods to solve several common problems with C++11.
C++0 stars
Explain
data_structure
Several commonly used data structures are implemented with C++11.
C++0 stars
Explain
machine_learning
Implement several common machine learning algorithms with sklearn.
Python0 stars
Explain
crawler
Several fun crawler cases implemented in Python.
Python0 stars
Explain