relyks's favorites | Hacker News

1.		Show HN: Ithihāsas – a character explorer for Hindu epics, built in a few hours (ithihasas.in)
		153 points by cvrajeesh 17 hours ago \| 39 comments
2.		If you started a company two years ago, many assumptions are no longer true (steveblank.com)
		194 points by tie-in 3 days ago \| 140 comments
3.		Repository Pattern with Hygienic Macros in Scheme – Lisp (jointhefreeworld.org)
		10 points by jjba23 4 days ago \| discuss
4.		Lean proved this program correct; then I found a bug (kirancodes.me)
		286 points by bumbledraven 12 hours ago \| 138 comments
5.		Introspective Diffusion Language Models (introspective-diffusion.github.io)
		91 points by zagwdt 4 hours ago \| 25 comments
6.		Against vibes: When is a generative model useful (williamjbowman.com)
		107 points by takira 34 days ago \| 28 comments
7.		Contextual commits – An open standard for capturing the why in Git history (vidimitrov.substack.com)
		37 points by vidimitrov 32 days ago \| 38 comments
8.		Reliable Software in the LLM Era (quint-lang.org)
		108 points by mempirate 33 days ago \| 34 comments
9.		The peril of laziness lost (dtrace.org)
		465 points by gpm 1 day ago \| 142 comments
10.		Pijul a FOSS distributed version control system (pijul.org)
		191 points by kouosi 7 days ago \| 30 comments
11.		Show HN: Pardonned.com – A searchable database of US Pardons
		496 points by vidluther 3 days ago \| 272 comments
12.		I still prefer MCP over skills (david.coffee)
		458 points by gmays 4 days ago \| 373 comments
13.		Research-Driven Agents: When an agent reads before it codes (skypilot.co)
		212 points by hopechong 4 days ago \| 54 comments
14.		How NASA built Artemis II’s fault-tolerant computer (acm.org)
		637 points by speckx 4 days ago \| 235 comments
15.		ML promises to be profoundly weird (aphyr.com)
		612 points by pabs3 5 days ago \| 602 comments
16.		MegaTrain: Full Precision Training of 100B+ Parameter LLMs on a Single GPU (arxiv.org)
		324 points by chrsw 6 days ago \| 57 comments
17.		Show HN: I built a tiny LLM to demystify how language models work (github.com/arman-bd)
		915 points by armanified 8 days ago \| 134 comments
18.		Writing Lisp is AI resistant and I'm sad (djhaskin.com)
		99 points by djha-skin 9 days ago \| 99 comments
19.		A case study in testing with 100+ Claude agents in parallel (imbue.com)
		68 points by thejash 10 days ago \| 54 comments
20.		Components of a Coding Agent (sebastianraschka.com)
		299 points by MindGods 9 days ago \| 90 comments
21.		Emotion concepts and their function in a large language model (anthropic.com)
		190 points by dnw 10 days ago \| 204 comments
22.		Functional programming accelerates agentic feature development (cyrusradfar.com)
		59 points by cyrusradfar 12 days ago \| 31 comments
23.		Embarrassingly simple self-distillation improves code generation (arxiv.org)
		656 points by Anon84 10 days ago \| 201 comments
24.		We replaced RAG with a virtual filesystem for our AI documentation assistant (mintlify.com)
		410 points by denssumesh 11 days ago \| 151 comments
25.		Category Theory Illustrated – Types (abuseofnotation.github.io)
		117 points by boris_m 11 days ago \| 22 comments
26.		[flagged] The future of code search is not regex – 100x faster than ripgrep (dmtrkovalenko.dev)
		60 points by neogoose 12 days ago \| 44 comments
27.		Slop is not necessarily the future (greptile.com)
		305 points by dakshgupta 13 days ago \| 484 comments
28.		I quit. The clankers won (dbushell.com)
		423 points by domysee 13 days ago \| 484 comments
29.		Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs (prismml.com)
		430 points by PrismML 13 days ago \| 153 comments
30.		TinyLoRA – Learning to Reason in 13 Parameters (arxiv.org)
		234 points by sorenjan 18 days ago \| 45 comments
		More

HN For You