Jędrzej Maczan

/ spelled as yenjay in English and 延杰 in Chinese! /

Solving 0/1 Knapsack problem with sliding window and Hirschberg algorithm [To be published]
Paged Out!, Issue #8, 2026

algorithms optimization math
torch-webgpu: PyTorch compiler and WebGPU runtime [In progress]
Open source

pytorch webgpu compilers
Use 20x less peak RAM with dp_knapsack_sliding_hirschberg, a new activation memory budget solver for PyTorch 2025-11-21
Open source

pytorch algorithms compilers optimization
Self-contained handwritten digit recognizer 2025-10-01
Paged Out!, Issue #7, page 11 · Open source

ml neural-networks from-scratch
A primer on Differentiable Architecture Search 2025-03-01
Paged Out!, Issue #6, page 5 · Open source

ml automl neural-networks pytorch
GPT in PyTorch 2024-12-01
Paged Out!, Issue #5, page 6 · Open source

pytorch llm transformers
Building automated machine learning with type inference 2024-06-01
Paged Out!, Issue #4, page 4 · Open source

automl ml python

I like computers!

Nowadays I focus on deep learning compilers and low level artificial intelligence systems, both research and building them from scratch
My fav tech: C++, Python, PyTorch, LLMs, ML compilers, CUDA, WebGPU, Linux, Rust and TypeScript
I'm a regular author for Paged Out! magazine, which is a cool nerdy zine for things like hacking, programming and everything computer-related
Night job:
- torch-webgpu - PyTorch compiler and WebGPU runtime
- tiny-vllm - High performance LLM inference engine, a younger sibling of vLLM, native C++ and CUDA implementation
- PyTorch - mainly compilers TorchDynamo and TorchInductor, less frequently as of 2026
Day job: Applying LLMs to things c:
GitHub: @jmaczan
Email: jedrzej@maczan.pl
Art of Anita: anitamaczan.pl, instagram: @anita_maczan
Other links (not very active):
- Mastodon: mathstodon.xyz/@jmaczan
- LinkedIn: click
- X/Twitter: @jedmaczan
- HackTheBox: @7563687575 (inactive, had fun some time ago)