yubblegum's favorites

1.		You already have a Git server (maurycyz.com)
		653 points by chmaynard 11 days ago \| 427 comments
2.		My trick for getting consistent classification from LLMs (verdik.substack.com)
		318 points by frenchmajesty 24 days ago \| 70 comments
3.		NanoChat – The best ChatGPT that $100 can buy (github.com/karpathy)
		1523 points by huseyinkeles 24 days ago \| 308 comments
4.		OpenZL: An open source format-aware compression framework (fb.com)
		434 points by terrelln 31 days ago \| 107 comments
5.		Designing agentic loops (simonwillison.net)
		284 points by simonw 37 days ago \| 117 comments
6.		John Coltrane's Tone Circle (roelsworld.eu)
		167 points by jim-jim-jim 64 days ago \| 64 comments
7.		Show HN: A zoomable, searchable archive of BYTE magazine (tsundoku.io)
		423 points by chromy 72 days ago \| 71 comments
8.		Scallop – A Language for Neurosymbolic Programming (scallop-lang.org)
		229 points by andsoitis 7 months ago \| 64 comments
9.		FFmpeg 8.0 adds Whisper support (ffmpeg.org)
		1033 points by rilawa 85 days ago \| 325 comments
10.		My 2.5 year old laptop can write Space Invaders in JavaScript now (GLM-4.5 Air) (simonwillison.net)
		577 points by simonw 3 months ago \| 396 comments
11.		Coding with LLMs in the summer of 2025 – an update (antirez.com)
		600 points by antirez 3 months ago \| 414 comments
12.		The Evolution of Caching Libraries in Go (maypok86.github.io)
		128 points by maypok86 4 months ago \| 34 comments
13.		The new skill in AI is not prompting, it's context engineering (philschmid.de)
		915 points by robotswantdata 4 months ago \| 518 comments
14.		[dupe] Embeddings are underrated (2024) (technicalwriting.dev)
		484 points by jxmorris12 5 months ago \| 150 comments
15.		Vision Now Available in Llama.cpp (github.com/ggml-org)
		550 points by redman25 6 months ago \| 104 comments
16.		June Huh dropped out to become a poet, now he’s won a Fields Medal (2022) (quantamagazine.org)
		219 points by bpierre 6 months ago \| 139 comments
17.		Mycoria is an open and secure overlay network that connects all participants (mycoria.org)
		336 points by doener 6 months ago \| 125 comments
18.		Local LLM inference – impressive but too hard to work with (medium.com/aazo11)
		84 points by aazo11 6 months ago \| 58 comments