AI product Open source
Handwired is a JavaScript tool that transforms ordinary programs containing conditionals, loops, and state into differentiable neural-network-style computations. It rewrites control flow into arithmetic using multiplication, addition, and ReLU, then collects those operations into a large matrix multiplication; the resulting program is designed to produce the same outputs without training. The computation can run in JavaScript or WebGPU, and the manually written weights remain differentiable for experiments such as training them or connecting the frozen program to another neural network. The repository includes live demos and can be run locally with Bun.
2 uses taken from transcripts — each links to the moment in the video.
Converts plain JavaScript into a neural network that needs no training; it rewrites if-else branches, loops, and state into simple arithmetic using only multiplication, addition, and ReLU so output exactly matches the original code, turning the whole program into one big matrix multiplication running in JavaScript or on WebGPU.
An open-source project by Chenglou that turns regular JavaScript into an exact neural net — no training, exact outputs, yet fully differentiable. It compiles if-else, loops and state into matrix multiplications and runs in the browser with no GPU.
2 in the library.