<aside> 💡 Notion Tip: Use this template to organize important information for your team. Add owners, verification, and tags to pages to keep them up to date. Just replace this sample content with your own.

</aside>

Who am I?

I am Drew, an Efficient-Gen-AI researcher and MLSys enthusiast.

Glad that you take interest in my wiki.

Attention

The Wiki will mostly be written in Chinese, sorry for those who use other languages.

Basics


Leet Code

Hot 100 Notes

Languages

Python

CPP

SHELL

Train


SFT

PyTorch Lightning

Megatron-LM

Transfomers Trainer

RLHF

VeRL

Both

LLaMA-Factory

Unsloth

Inference


Engine

vLLM

SGLang

Quantization

AWQ

Efficient Kernels

Flash Attention

Easy Kernel Lang

Triton

TileLang

CUDA

CUDA

CUTLASS

Algorithm


Decoding

Speculative Decoding

无标题

KV Cache

KV Cache Compression

Prefilling

Sparse Attention

HPC


ML


OS