Lukas Fisch

nanoRL. Python

jepa. implementing minimal versions of joint-embedding predictive architecture (JEPA)

tinyworlds. A minimal implementation of DeepMind's Genie world model

CUDA-Agent. CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation

ANE. Training neural networks on Apple Neural Engine via reverse-engineered private APIs

canvas_ity. A tiny, single-header <canvas>-like 2D rasterizer for C++

zig-minimal-kernel-x86. Minimal x86 Kernel - built in Zig

BarraCUDA. Open-source CUDA, Triton and HIP compiler targeting multiple GPU and CPU architectures.

plutosvg. Tiny SVG rendering library in C

coi. A type-safe, component-based language for building reactive WASM web apps.

regular-express. A completely visual regular expression tool. Build and edit complex regular expressions without ever worrying about syntax. Currently in early alpha.

meetily. Privacy first, AI meeting assistant with 4x faster Parakeet/Whisper live transcription, speaker diarization, and Ollama summarization built on Rust. 100% local processing. no cloud required. Meetily (Meetly Ai - https://meetily.ai) is the #1 Self-hosted, Open-source Ai meeting note taker for macOS & Windows. Understand How to write meeting minutes

glText. Cross-platform single header text rendering library for OpenGL

Astral. x86-64 Operating System

arTui. A small TUI to read and track recent arXiv papers

datoviz. ⚡ Datoviz: high-performance GPU rendering for scientific data visualization

braincraft. Braincraft challenge — 1000 neurons, 100 seconds, 10 runs, 2 choices, no reward

mojo-gpu-puzzles. Learn GPU Programming in Mojo🔥 by Solving Puzzles

modded-nanogpt. NanoGPT (124M) in 90 seconds

tinygrad. You like pytorch? You like micrograd? You love tinygrad! ❤️

cleanrl. High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

PufferLib. Puffing up reinforcement learning

T1Prep. T1Prep: T1 PREProcessing pipeline (aka PyCat)