Hacker News: rdos

New comment by rdos in "I'm Tired of Talking to AI"

rdos — Wed, 27 May 2026 13:56:48 +0000

> Recently someone messaged me on Reddit about my post. I replied. They wrote again, I replied again. After a few messages I realized I was talking to an AI agent.

My exact experience. The irony was that we were talking about AI agents

New comment by rdos in "Technology Radar Volume 34 – April 2026"

rdos — Wed, 15 Apr 2026 15:17:44 +0000

The Tools detail page is wrong, it's the one from last release, November 2025.

New comment by rdos in "Picasso’s Guernica (Gigapixel)"

rdos — Wed, 15 Apr 2026 14:09:34 +0000

almost fell for it

New comment by rdos in "Pretty Fish: A better mermaid diagram editor"

rdos — Wed, 15 Apr 2026 13:38:37 +0000

I can't seem to change the colors of the pie chart, other than the predefined themes. But all of those are horrible for a pie chart.

New comment by rdos in "Claude mixes up who said what"

rdos — Thu, 09 Apr 2026 14:55:42 +0000

> This bug is categorically distinct from hallucinations or missing permission boundaries

I was expecting some kind of explanation for this

New comment by rdos in "Quantization from the Ground Up"

rdos — Wed, 08 Apr 2026 09:22:39 +0000

Sure, two 3060 can pull usable performance on an usable LLM, but a single one can't (yet).

> 3x RTX 3060 less tgab the price of a 3090

Interesting, here it is around the same. 200-250€ for a used 12GB 3060 and 600-800 for a used 3090€.

New comment by rdos in "Salomi, a research repo on extreme low-bit transformer quantization"

rdos — Thu, 02 Apr 2026 10:47:52 +0000

Was any text in the repo NOT written by AI?

New comment by rdos in "Quantization from the Ground Up"

rdos — Thu, 26 Mar 2026 10:46:06 +0000

14B even at Q4 isn't realistic for coding on a single 12GB RTX 3060. Token speed is too slow. After all they are dense models. You aren't getting a good MoE model under 30B. You can do OCR, STT, TTS really well and for LLMs, good use cases are classification, summarization and extraction with <10B models.

New comment by rdos in "My Journey to a reliable and enjoyable locally hosted voice assistant (2025)"

rdos — Tue, 17 Mar 2026 14:56:19 +0000

> llama.cpp (previously Ollama)

I almost fainted

New comment by rdos in "GLM-OCR – A multimodal OCR model for complex document understanding"

rdos — Wed, 11 Feb 2026 16:26:02 +0000

Interesting. Won't stuff like entity extraction suffer? Especially in multilingual use cases. My worry is that a smaller model might not realize some text is actually a persons name because it is very unusual.

New comment by rdos in "GLM-OCR – A multimodal OCR model for complex document understanding"

rdos — Wed, 11 Feb 2026 16:11:28 +0000

Is it possible for such a small model to outperform gemini 3 or is this a case of benchmarks not showing the reality? I would love to be hopeful, but so far an open source model was never better than a closed one even when benchmarks were showing that.

New comment by rdos in "Task-free intelligence testing of LLMs"

rdos — Fri, 09 Jan 2026 10:38:49 +0000

This is very interesting. Especially the last part where it shows gpt-5.2 and gpt-oss and their very similar and unique outcome of being 90%+ Serious.

I tested this locally and got the same result with gpt-oss 120b. But only on the default 'medium' reasoning effort. When I used 'low' I kept getting more playful responses with emojis and when I used 'high' I kept getting more guessing responses.

I had a lot of fun with this and it provided me with more insight than I would have thought.

New comment by rdos in "Opus 4.5 is not the normal AI agent experience that I have had thus far"

rdos — Wed, 07 Jan 2026 15:53:19 +0000

> Are you saying this from experience?

Yes. I mostly work on Quarkus microservices and use cursor with auto agent mode.

> we wouldn't give an AI some vague requirements and ask it to build something > we would discuss as a team

seems like a reasonable workflow. It's the polar opposite of what was written in the blog post. That is the usual, easy way people use agents and what I think is the wrong path. May I also ask what language and/or framework you work with where so much context works good enough?

> Asking AI to explain code and help me learn how it works means I can pick up new systems significantly quicker.

Summarization is generaly a great task for LLMs

New comment by rdos in "Why the trans flag emoji is the 5-codepoint sequence it is"

rdos — Wed, 07 Jan 2026 15:24:26 +0000

I didn't read the blog yet because I clicked on cat pics and there weren't any!!!

New comment by rdos in "Opus 4.5 is not the normal AI agent experience that I have had thus far"

rdos — Wed, 07 Jan 2026 09:59:09 +0000

LLM's are good at making stuff from scratch and perfect when you don't have to worry about the codes future. 'Research' can be a great tool. But LLMs are horrible in big codebases and multiple micro services. Also at making decision, never let it make a decision for you. You need to know what's happening and you can't ship straight AI code. It can save time, but it's not a lot and it won't replace anyone.

New comment by rdos in "Ask HN: What Are You Working On? (December 2025)"

rdos — Mon, 15 Dec 2025 10:27:42 +0000

I don't want to sound rude, but what was your reason to go from scratch instead of joining an already established, open source effort? The likes of Cline, Roo, Continue, ...

New comment by rdos in "Ask HN: What Are You Working On? (December 2025)"

rdos — Mon, 15 Dec 2025 10:25:26 +0000

nice, what's your approach? Graphs?

New comment by rdos in "Programming peaked"

rdos — Thu, 04 Dec 2025 13:58:54 +0000

this is going straight into my funny folder

New comment by rdos in "Advent of Code 2025"

rdos — Mon, 01 Dec 2025 09:15:05 +0000

It won't load for me right now

New comment by rdos in "I have recordings proving Coinbase knew about breach months before disclosure"

rdos — Mon, 17 Nov 2025 12:29:47 +0000

Hello, would you add something to this list? I think it's pretty good

> Over‑polished prose – flawless grammar, overly formal tone, and excessive wordiness.

> Repetitive buzzwords – phrases like “delve into,” “navigate,” “vibrant,” “comprehensive,” etc.

> Lack of perspective shifts – AI usually sticks to a single narrative voice; humans naturally mix first, second, and third person.

> Excessive em‑dashes – AI tends to over‑use them, breaking flow.

> Anodyne, neutral stance – AI avoids strong opinions, trying to please every reader.

> Human writing often contains minor errors, idiosyncratic punctuation, and a more nuanced, opinionated voice.

> It's not just x, it's y