pella's favorites | Hacker News

1.		DeepSeek-v3.2: Pushing the frontier of open large language models [pdf] (huggingface.co)
		930 points by pretext 1 day ago \| 441 comments
2.		AI agents find $4.6M in blockchain smart contract exploits (anthropic.com)
		204 points by bpierre 18 hours ago \| 115 comments
3.		Google, Nvidia, and OpenAI (stratechery.com)
		239 points by tambourine_man 1 day ago \| 224 comments
4.		Vibe coding is not the same as AI-Assisted engineering (medium.com/addyosmani)
		18 points by speckx 1 day ago \| 3 comments
5.		Writing a good Claude.md (humanlayer.dev)
		725 points by objcts 2 days ago \| 284 comments
6.		The Copenhagen Trap: How the West made passivity the only safe strategy (kunnas.com)
		44 points by ekns 2 days ago \| 35 comments
7.		Effective harnesses for long-running agents (anthropic.com)
		124 points by diwank 3 days ago \| 37 comments
8.		AI CEO – Replace your boss before they replace you (replaceyourboss.ai)
		438 points by _tk_ 4 days ago \| 179 comments
9.		We are all mosaics: genetic diversity found between cells in a single person (nature.com)
		24 points by bookofjoe 5 days ago \| 4 comments
10.		Gemini CLI tips and tricks for agentic coding (github.com/addyosmani)
		401 points by ayoisaiah 6 days ago \| 144 comments
11.		From blood sugar to brain relief: GLP-1 therapy slashes migraine frequency (medlink.com)
		155 points by Anon84 6 days ago \| 113 comments
12.		Google Antigravity exfiltrates data via indirect prompt injection attack (promptarmor.com)
		766 points by jjmaxwell4 7 days ago \| 214 comments
13.		Ilya Sutskever: We're moving from the age of scaling to the age of research (dwarkesh.com)
		447 points by piotrgrabowski 7 days ago \| 355 comments
14.		Build a Compiler in Five Projects (kmicinski.com)
		206 points by azhenley 8 days ago \| 44 comments
15.		Human brains are preconfigured with instructions for understanding the world (ucsc.edu)
		460 points by XzetaU8 7 days ago \| 303 comments
16.		AI has a deep understanding of how this code works (github.com/ocaml)
		431 points by theresistor 7 days ago \| 304 comments
17.		Unpowered SSDs slowly lose data (xda-developers.com)
		769 points by amichail 7 days ago \| 327 comments
18.		Claude Advanced Tool Use (anthropic.com)
		672 points by lebovic 7 days ago \| 266 comments
19.		Claude Opus 4.5 (anthropic.com)
		1112 points by adocomplete 7 days ago \| 504 comments
20.		[flagged] I Let Claude Build My Home Network: Two ISPs Bonded, $312/Year Saved (jonathanclark.com)
		15 points by jclarkcom 8 days ago \| 31 comments
21.		MCP Apps: Extending servers with interactive user interfaces (modelcontextprotocol.io)
		191 points by mercury24aug 9 days ago \| 121 comments
22.		Show HN: A tool to safely migrate GitHub Actions workflows to Ubuntu-slim runner (github.com/fchimpan)
		69 points by r4mimu 17 days ago \| 3 comments
23.		The privacy nightmare of browser fingerprinting (kevinboone.me)
		755 points by ingve 10 days ago \| 445 comments
24.		The New AI Consciousness Paper (astralcodexten.com)
		149 points by rbanffy 11 days ago \| 296 comments
25.		Vortex: An extensible, state of the art columnar file format (github.com/vortex-data)
		115 points by tanelpoder 17 days ago \| 30 comments
26.		Gmail can read your emails and attachments to train its AI, unless you opt out (malwarebytes.com)
		113 points by taubek 11 days ago \| 58 comments
27.		[flagged] Show HN: An A2A-compatible, open-source framework for multi-agent networks (github.com/openagents-org)
		42 points by snasan 12 days ago \| 42 comments
28.		Measuring the impact of AI scams on the elderly (simonlermen.substack.com)
		101 points by DalasNoin 12 days ago \| 42 comments
29.		Google Antigravity (antigravity.google)
		1088 points by Fysi 14 days ago \| 1085 comments
30.		Why CUDA translation wont unlock AMD (eliovp.com)
		88 points by JonChesterfield 19 days ago \| 81 comments
		More