Hacker News Reader

v3ss0non Devvm.dev: One VM for every project

why this when we have mature Containerized or VM system like Incus and LXC?

v3ss0non Ask HN: What's your current AI agentic setup?

DSH + 3xSpark + GLM5.3 Flash

I dont think i would need anything anymore. Its doing alot more than what i did 2 years ago in within 2 months.

v3ss0non Writing seen in PHerc. 306B, for the first time in 2000 years

Pherc 306B A12B when?

v3ss0non [dead]

looks like he vibe code and tell it go post HN

v3ss0non [dead]

LOL why u even need LLM to write that one line of link?

v3ss0non Python's experimental JIT to be removed

Guido thought about it before he get into Microsoft.

v3ss0non Early progress on accelerating drug development with Claude

Which model you used? Same for me too , until GML 5.3 (flash) hosted locally. GLM 5.3 is very resistant to Hallucination and Gibberish.

v3ss0non Enlightenment 0.28.0 Release

I started trying like 10 years ago , it looks good but never managed to get it to stable. But forgivable because all the WMs , including gnome (was stable at that time to some point) , and KDE , same as crash heavy as that .

But then KDE become stable so switch to it.

Then Last time i tried to give it a chance was 2022 STILL Same broken WM with looks good stuff and they are trying to build it like KDE at that point including multiple apps.

At that point Enlightenment no longer looks good , KDE had caught up all the compositor effects and looking cool , can configure as much as we want. Elightment at that point still crashy as hell and never looked back.

v3ss0non Enlightenment 0.28.0 Release

I can't believe this crash ridden peice of WM Still exist.. It was horror just to setup something.

v3ss0non Qwenputer: A functioning, deterministic RISC-V CPU inside Qwen3.5

And -- that is our reality.

v3ss0non Early progress on accelerating drug development with Claude

on average it is fine. On difficult tasks it still does hallucinate. - like decision based on huge policy documents ( context reaching around 900k)

A lot less these days tho.

v3ss0non Early progress on accelerating drug development with Claude

Yo Dawg , you love hallucination so we develop drugs with LLM.

v3ss0non Python's experimental JIT to be removed

Onlything is a lot of cextensions are super slow there. that is only bottleneck left which could be solved with resources that those guys spent on experimental jit.

v3ss0non Python's experimental JIT to be removed

They should had just port everything to pyPy.

v3ss0non The Mythical Man-Month in the Age of AI

Git Worktress, Sub Agent WIth Team Leader agents. I get things done alot faster, You are just doing it wrong.

v3ss0non Microsoft Decision-1: 9B Qwen Model Launch Explained

https://news.ycombinator.com/threads?id=v3ss0n#:~:text=alrea...

Sorry i reply to wrong comment.

v3ss0non Microsoft Decision-1: 9B Qwen Model Launch Explained

already , i self hosted Deekseek 4.1 Flash , Qwen 3.8 Flash Next , Glm 5.3 flash , tested it , all benchmaxxed to the point that all of them can draw that well , much better than other logos .

v3ss0non Microsoft Decision-1: 9B Qwen Model Launch Explained

Sorry I didn't knew it was started here and , lol I didn't know it was viralized by Simon of django! Now these days , many LLMs are benchmaxxed to it.

v3ss0non Facebook Pages require paid subscription to post more than two links per month

I hated react and don't use at all but credit to credit due

v3ss0non Facebook Pages require paid subscription to post more than two links per month

PyTorch .. React ..

That two libraries already build Frontend and AI.

v3ss0non Ultramarine 1 Flash

Me too Deekseek harness , is sleek , solid , and flexible.

v3ss0non OpenCode 2.0

Try pangolin for tunneing , it works so well.

v3ss0non OpenAI engineer arrested in bugging of executive conference room

FBI!! OPEN UP!

v3ss0non Total Annihilation, rebuilt for modern computers

I have been in 50 v 50 .. and that is mervelous .. none of the AAA Game company can do a RTS SMOOTH as butter at 140 FPS while whole map is filled with 100 player explosions.

v3ss0non Doory – Open-Source Canva

Vibe coded or not , Software quality comes with proper maintenance and engineering principles.. i dont believe full vibers will do those.

v3ss0non Doory – Open-Source Canva

Thats the point. They wont care about the craft , design decision ,best engineering princinples.

v3ss0non Doory – Open-Source Canva

Yeah ... looks like that is the answer these days.

v3ss0non Doory – Open-Source Canva

Vibe coded or not , proper engineering , testing , fixing , iteration is still needed . Your don't have one , full of crashes and bugs.

v3ss0non Doory – Open-Source Canva

fully vibed low effort clone atually. it is now crashing while testing.

v3ss0non Doory – Open-Source Canva

So with this future of vibe coded tools , how to choose which one to use?