Show HN: Watch weight-space programming in action with WebGPU
Sentiment Mix
Expert Signals
wsxiaoys
author • 1 mention
Hacker News
source • 1 mention
AI-Generated Claims
Generated from linked receipts; click sources for full context.
I wanted to build a full weight-space programming loop entirely in the browser with WebGPU: let an LLM inspect Qwen3.5-0.8B, edit its weights, evaluate, and repeat.the full loop needed too much VRAM.
Supported by 1 story
so I turned it into an interactive walkthrough of a concrete example: editing 0.016% of a model's weights to reduce reasoning loops, no gradient descent.Based on exp8 of "Sparks of Weight-Space Programming.": https://news.ycombinator.com/item?id=50341174
Supported by 1 story
Paper to Product Links
Related Events
Show HN: IT Thinks – Build a GPU from a Single NAND Gate
Hardware • 11/20/2026
Weight-Space Programming builds neural networks like a 1990s researcher
Uncategorized • 11/20/2026
Show HN: Soccer-Bench – Watch teams of LLM agents play 5v5 soccer
LLMs • 11/20/2026
Launch HN: Onlook (YC W25) – An open-source developer tool for designers
Open Source • 11/20/2026
Chrome AI Features Wasting Much Bandwidth
Uncategorized • 11/20/2026