Lean Grimoire(terrytao.wordpress.com)
51 points by smilelamp | 8 comments
Terence Tao blog post announcing Lean Grimoire, a coding-agent-rebuilt version of his QED logic crafting game with Lean proof display and a fantasy mode.
First screen, captured . Open the video.
Full page
Captured . Click for the whole screenshot.
Looks similar
Pigshell: Unix the Web(pigshell.com)
63 points by pbohun | 18 comments
Pigshell 0.6.4 is a browser-based Unix-style shell that represents web resources and cloud data as files you can process with command pipelines.
Variable-width Bézier curves (2025)(luthert.web.illinois.edu)
26 points by leduyquang753 | 7 comments
A blog post by Luther Tychonievich explaining how to draw variable-width (pressure-sensitive) Bézier curves with current 2D graphics APIs for board-work tablet apps.
Awesome repositories are no longer awesome(erikgahner.github.io)
239 points by erikgahner | 116 comments
A blog post arguing that GitHub 'awesome' curated lists have become less useful due to outdated content and floods of AI-generated pull requests promoting vibe-coded sites.
Pliny – Letters – Book 7 Letter 27 (On Ghosts)(vroma.org)
8 points by yubblegum | 1 comments
English translation of Pliny the Younger's Letter 7.27 to Sura, recounting ghost stories including Athenodorus and the haunted house in Athens.
The History of Djgpp (1998)(delorie.com)
59 points by atan2 | 23 comments
DJ Delorie's 1998 history of DJGPP, the port of GCC to MS-DOS, covering its origins, the go32 extender, and releases through version 2.02.
Related stories
A 30B Qwen model walks into a Raspberry Pi and runs in real time(byteshape.com)
338 points by dataminer | 121 comments
ByteShape blog post describing how they quantized Qwen3-30B-A3B-Instruct with their Shapelearn bitlength method to run in real time on a Raspberry Pi.
Can a Linear Program Do Weight-Space Programming?(nathan.rs)
285 points by emschwartz | 41 comments
Blog post arguing that a model's weight edit fixing Qwen3.5-0.8B thinking loops can be reproduced by solving a linear program, with fewer changed weights.
Detecting model weight tampering in GPU memory (Rome attacks)(stealthium.io)
15 points by rafaeldavidtin | 1 comments
Blog post from Stealthium on detecting tampering of AI model weights in GPU memory, such as ROME model-editing attacks.
Using AI to write better code more slowly(nolanlawson.com)
1102 points by signa11 | 407 comments
Blog post by Nolan Lawson arguing that LLMs can be used to write higher-quality code more slowly, such as finding bugs, rather than producing low-quality code fast.
Running Qwen3.8 at 255 tok/s on a single 5090(meshapplied.com)
7 points by Ir0n | 0 comments
A page from meshapplied.com about running a Qwen LLM at 255 tokens per second on a single Nvidia RTX 5090 GPU; no page content was captured.
Astra can read and write weights to improve Qwen(rikiyat.github.io)
2884 points by dbreunig | 869 comments
MIT CSAIL blog post reporting that frontier LLMs can read, fix, and write neural network weights directly, including editing Qwen3.5-0.8B to improve math performance.