Hacker News: orsorna

New comment by orsorna in "Our $445M Series D"

orsorna — Fri, 09 Oct 2026 13:57:39 +0000

Right...so best interview process in the game for employers, not so much employees.

New comment by orsorna in "We have LLMs now. Why are the docs still wrong?"

orsorna — Fri, 09 Oct 2026 01:03:01 +0000

This is very curious to me. When you resolve a domain for the first time, does your resolver look up this kind of metadata for the domain before completing resolution?

New comment by orsorna in "Worth Building"

orsorna — Tue, 06 Oct 2026 00:11:49 +0000

It is partially the result from chilling effects of hosting AI generated code on your public portfolio. Not worth the ire.

New comment by orsorna in "The designer lamps in my house"

orsorna — Mon, 05 Oct 2026 18:30:36 +0000

No because you likely do not drench yourself in hubris daily.

New comment by orsorna in "Pi 1.0"

orsorna — Fri, 02 Oct 2026 00:03:16 +0000

I am wondering what use people get out of this. What I like about pi is that, using proprietary harnesses to begin with, I already somewhat know what kind of features I use and features I wish I could use. Asking pi to self modify itself (with a smart enough model) is a one time cost for a fully vetted feature that I personally approve. Do users of OMP find all of its additions useful?

New comment by orsorna in "MongoDB CEO resigns to join Meta"

orsorna — Mon, 28 Sep 2026 15:48:09 +0000

>He's never going to be hired as CEO by anybody for the rest of his career

I'd eat my shorts if this was true. But even then he is likely set for life with his wealth, pre Meta offer.

New comment by orsorna in "OpenAI Feared "Optics" of what might appear on Hacker News"

orsorna — Sun, 27 Sep 2026 13:59:10 +0000

Fallacy that only serves corporate interests, as he points out.

New comment by orsorna in "Jev in 25 Lines of Python"

orsorna — Wed, 23 Sep 2026 11:59:03 +0000

Oh, I thought it was some weird branding thing. So then I looked it up.

"System One" and "System Two" were coined in some pop science book...so back to its usage being a marketing ploy.

New comment by orsorna in "Does an open-weight decision model beat a hosted one? Jev vs. Laya"

orsorna — Wed, 23 Sep 2026 02:36:24 +0000

For anyone who was piqued by the title but couldn't stomach the unedited LLM blogpost, you can read the similarity AI-genned docs for Laya directly: https://github.com/NandhaKishorM/laya

New comment by orsorna in "The Economics of Open-Weight Inference"

orsorna — Tue, 22 Sep 2026 18:19:38 +0000

It was a breath of fresh air for me because it was nice going to a model that is "smart enough".

New comment by orsorna in "AI Has No Wisdom and Neither Will You"

orsorna — Tue, 22 Sep 2026 13:13:56 +0000

>The trillion dollar question is how you do this, if your employees do not care (they are optimising for salary & time spent not code quality)

Sounds like there is a compensation problem then.

New comment by orsorna in "M5 Ultra Mac Studio Review"

orsorna — Mon, 21 Sep 2026 19:22:31 +0000

You definitely need more RAM if you are not satisfied with small context windows, especially if the weights take a large % of the total memory to boot.

New comment by orsorna in "M5 Ultra Mac Studio Review"

orsorna — Mon, 21 Sep 2026 16:31:59 +0000

Is it that silly? You could run multiple 27B models in parallel.

New comment by orsorna in "ZuckOff is a free app that sees Meta glasses before they see you"

orsorna — Mon, 21 Sep 2026 14:03:10 +0000

Context https://www.businessinsider.com/mark-zuckerberg-used-a-boost...

New comment by orsorna in "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint"

orsorna — Fri, 18 Sep 2026 04:03:52 +0000

I don't do anything fancier than inference, and I only use llama.cpp, which supports rOCM. I've had few issues; most GGUFs I download work right out of the box. Nearly any popular model has a quant that just works. But as you can see I don't use my GPU for anything weird or nonstandard.

New comment by orsorna in "Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint"

orsorna — Fri, 18 Sep 2026 00:18:46 +0000

7900 XT is a sleeper card. When I initially bought it, it was priced at the lowest wattage per $ per GB VRAM (not normalized for token speeds...) Although I ended up swapping for the XTX because that 4GB means everything in just increasing the context window. At 8bit KV my window is over 200k, and although qwen3.8 loves vomiting out tokens as part of its reasoning chain I trust it enough to get assigned tasks done eventually, which I could not say of any model before its release.

New comment by orsorna in "My AI agent gets his own calendar"

orsorna — Thu, 17 Sep 2026 13:25:15 +0000

This is just blogspam, right? The author even brags about hitting the HN front page in such a short amount of time, in a previous post. All he talks about is his agent (that we are supposed to "know"? arrogant). Not a lot of discussion from him about his results.

New comment by orsorna in "Ask HN: How can DigitalOcean not afford $50, but can afford $3M to Omarchy?"

orsorna — Sun, 13 Sep 2026 04:09:48 +0000

All signs point to these donations being strictly ideologically motivated, and not because of the alleged merits Omarchy totes. I could further argue how sad it is that efforts like these receive tons of money versus struggling FOSS projects, all for a strictly non-technical reason.

New comment by orsorna in "Jellyfin 12.0"

orsorna — Tue, 08 Sep 2026 12:25:33 +0000

There is also an open source tvOS client called Swiftfin. The experience isn't as good though.

You can also transcode your files to formats that the free version of Infuse supports if you are a miser.

New comment by orsorna in "An Alien Mind"

orsorna — Mon, 07 Sep 2026 16:28:08 +0000

What investor concerns prompted writing this one?