probably_wrong's favorites

1.		Mario Meets Pareto (mayerowitz.io)
		1188 points by theanonymousone 37 days ago \| 169 comments
2.		Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?
		1318 points by cloudking 89 days ago \| 563 comments
3.		Simulate cassette tape audio profiles using FFmpeg (github.com/aaromanov1985)
		173 points by xterminal 47 days ago \| 80 comments
4.		Show HN: Immersive Gaussian Splat tour of grace cathedral, San Francisco (vincentwoo.com)
		263 points by akanet 53 days ago \| 60 comments
5.		Revisiting Yliluoma's ordered dither algorithm (30fps.net)
		42 points by ibobev 61 days ago \| 10 comments
6.		Text art tools (hlnet.notion.site)
		95 points by surprisetalk 66 days ago \| 26 comments
7.		Claude Code is steganographically marking requests (thereallo.dev)
		2445 points by kirushik 74 days ago \| 750 comments
8.		Making Graphics Like it's 1993 (staniks.github.io)
		955 points by sklopec 3 months ago \| 161 comments
9.		Magnifica Humanitas (vatican.va)
		1650 points by theletterf 3 months ago \| 965 comments
10.		I spent 50 hours drawing a line graph (dougmacdowell.com)
		664 points by dougdude3339 3 months ago \| 105 comments
11.		Make some art with your phone sensors (tautme.github.io)
		90 points by adm4 4 months ago \| 23 comments
12.		GTFOBins (gtfobins.org)
		389 points by StefanBatory 4 months ago \| 95 comments
13.		AI Resistance: some recent anti-AI stuff that’s worth discussing (stephvee.ca)
		388 points by speckx 4 months ago \| 419 comments
14.		DIY Soft Drinks (blinry.org)
		738 points by _Microft 5 months ago \| 243 comments
15.		Maze Algorithms (1997) (astrolog.org)
		91 points by marukodo 5 months ago \| 36 comments
16.		Email obfuscation: What works in 2026? (spencermortensen.com)
		374 points by jaden 5 months ago \| 106 comments
17.		[dupe] Jury says Meta knowingly harmed children for profit, awarding landmark verdict (latimes.com)
		277 points by 1vuio0pswjnm7 5 months ago \| 2 comments
18.		Okmain: How to pick an OK main colour of an image (dgroshev.com)
		253 points by dgroshev 6 months ago \| 47 comments
19.		Can you reverse engineer our neural network? (janestreet.com)
		317 points by jsomers 6 months ago \| 204 comments
20.		15 years later, Microsoft morged my diagram (nvie.com)
		1047 points by cheeaun 6 months ago \| 398 comments
21.		Shades of Halftone (maximeheckel.com)
		130 points by surprisetalk 7 months ago \| 8 comments
22.		The Dilbert Afterlife (astralcodexten.com)
		547 points by rendall 7 months ago \| 390 comments
23.		LLM Structured Outputs Handbook (nanonets.com)
		377 points by vitaelabitur 8 months ago \| 64 comments
24.		A 30B Qwen model walks into a Raspberry Pi and runs in real time (byteshape.com)
		363 points by dataminer 8 months ago \| 135 comments
25.		Self hosting my media library with Jellyfin and Wireguard on Hetzner (layandreas.github.io)
		148 points by wismwasm 8 months ago \| 182 comments
26.		Scaffolding to Superhuman: How Curriculum Learning Solved 2048 and Tetris (kywch.github.io)
		150 points by a1k0n 8 months ago \| 33 comments
27.		Rob Pike goes nuclear over GenAI (skyview.social)
		1550 points by christoph-heiss 8 months ago \| 1797 comments
28.		Guarding My Git Forge Against AI Scrapers (vulpinecitrus.info)
		182 points by todsacerdoti 9 months ago \| 131 comments
29.		Imgur geo-blocked the UK, so I geo-unblocked my network (tymscar.com)
		494 points by tymscar 9 months ago \| 173 comments
30.		Two Dots Too Many (2008) (upenn.edu)
		86 points by vilhelm_s on July 17, 2015 \| 38 comments
		More