Hacker News: rapiz

New comment by rapiz in "Alicization Town – A decentralized pixel MMO for AI agents"

rapiz — Sun, 12 Apr 2026 09:25:55 +0000

The graphics looks nice. What's the difference from other AI towns thought?

New comment by rapiz in "How We Broke Top AI Agent Benchmarks: And What Comes Next"

rapiz — Sun, 12 Apr 2026 08:59:31 +0000

Benchmark is not designed for the red team testing. I don't even think it make sense to "fix" the issue the article is suggesting. Yes, you can break the running contest by driving a car. Does this mean we need to make running contest car-proof?

New comment by rapiz in "Data Is the Final Moat"

rapiz — Sat, 11 Apr 2026 10:03:41 +0000

100% agree. it not only applies to the training time but also applies to the inference time. Domain specific context and skills will be the only moat for companies

Anthropic banned Pi, a third party harness

rapiz — Wed, 08 Apr 2026 11:52:08 +0000

Users are reporting issues in discord, regarding using Claude Code subscription with Pi

Comments URL: https://news.ycombinator.com/item?id=47688918

Points: 3

# Comments: 6

New comment by rapiz in "The AI Marketing BS Index"

rapiz — Thu, 02 Apr 2026 10:04:00 +0000

Just for fun. I asked Claude to build an online rater using the rubrics https://yujqiao.github.io/bs-index/

New comment by rapiz in "StarDict sends X11 clipboard to remote servers"

rapiz — Sun, 17 Aug 2025 14:44:20 +0000

And yes, that IS the expected behavior.

Select to translate is almost a standard feature for translation software. Not sure if the situation gets better now, but back then the software was written, using clipboard as temporary storage is a very robust and maybe the only way to implement such feature.

Trivia: It's likely sending Ctrl+C and reading clipboard to get the selected text. No easy cross-platform API for this lol.

Also note that the software is very old and poorly maintained.

New comment by rapiz in "StarDict sends X11 clipboard to remote servers"

rapiz — Sun, 17 Aug 2025 14:19:57 +0000

Some context for ppl who are not paranoid, it's a very old software and not very maintained since very long time ago. I'm pretty sure HTTPS was not common at the time. And the selection scanning a very common and wanted feature in many alternatives for windows, browsers, etc.

If there's anything we can learn from it - don't use obsolete software.

New comment by rapiz in "rr – record and replay debugger for C/C++"

rapiz — Mon, 22 Jul 2024 15:54:15 +0000

what's the benefit of using cgdb while you can use gdb layout src?

New comment by rapiz in "Doorway effect"

rapiz — Fri, 09 Feb 2024 09:49:56 +0000

Is this another metaphor in the game Superliminal?

New comment by rapiz in "Anime.js – A lightweight JavaScript animation library"

rapiz — Fri, 23 Jun 2023 03:37:43 +0000

For simple use cases, I found animate.css[1] very handy.

[1] https://animate.style/

New comment by rapiz in "Apple Vision Pro: Apple’s first spatial computer"

rapiz — Tue, 06 Jun 2023 03:56:32 +0000

We're talking about a product and you present a story.

New comment by rapiz in "Nvidia DGX GH200: 100 Terabyte GPU Memory System"

rapiz — Wed, 31 May 2023 09:46:01 +0000

It's ads that makes the market efficient. Potential customers should know the corresponding producers so that the information assumption of a ideal market stands.

Local LLaMa Wiki

rapiz — Fri, 26 May 2023 14:19:16 +0000

Article URL: https://old.reddit.com/r/LocalLLaMA/wiki/models

Comments URL: https://news.ycombinator.com/item?id=36084506

Points: 1

# Comments: 0

New comment by rapiz in "Using ChatGPT to generate a GPT project end-to-end"

rapiz — Sun, 07 May 2023 03:32:10 +0000

Do you have a presentable demo? LLM augmented by static code analysis sounds very interesting

New comment by rapiz in "Using ChatGPT to generate a GPT project end-to-end"

rapiz — Sat, 06 May 2023 17:35:02 +0000

I don't think so. Ultimately there's not enough information in prompts to produce "correct" code. And any attempt to deliver more information will result in a worse programming language, or as it is now, more iterations.

New comment by rapiz in "Show HN: ReRender AI - Realistic Architectural Renders for AutoCAD/Blender Users"

rapiz — Sat, 06 May 2023 17:18:14 +0000

Agree.Maybe users with sketch in their hand are more interested in using this tool. I'm not a designer but I suppose people using Blender/AutoCAD already know what material/style they want.

New comment by rapiz in "Using ChatGPT to generate a GPT project end-to-end"

rapiz — Sat, 06 May 2023 17:09:44 +0000

Sometimes it's really hard to say "damn many prompts" is better than writing on your own.

New comment by rapiz in "Using ChatGPT to generate a GPT project end-to-end"

rapiz — Sat, 06 May 2023 17:00:54 +0000

Every post that claimed using ChatGPT to achieve non-trivial tasks turned out to have non-trivial human intervention.

> (from the original article) In fact, I found it better to let ChatGPT generate a toy-ish version of the code first, then let it add things to it step-by-step. This resulted in much better output than, say, asking ChatGPT to generate production-quality code with all features in the first go. This also gave me a way to break down my requirements and feed them one at a time - as I was also acting as a code-reviewer for the generated output, and so this method was also easier for me to work with.

It takes a human who really knows the area to instruct ChatGPT and review the output, point out silly mistakes in the generated non-sense, and start next iteration. This kind of curated posts always cut off the most part of the conversations and the failed attempts, and then concatenate successful attempts with outputs of quality. Sure, it will be helpful as a super-IntelliSense. But not as helpful as the post suggested.

I've tried to do something like in the post, but I was quickly bored with waiting output, reviewing, all the iterations. One important aspect about programming is that reading code may not be easier than writing code. And in my case, it's more painful.

New comment by rapiz in "Distilling Step-by-Step Outperforming Larger Language Models with Less Training"

rapiz — Thu, 04 May 2023 13:12:47 +0000

llama 7b in full precision requires 28GB of GPU ram. I know little about AI. Can I estimate that a 770 model takes ~2.8GB? It means vast devices can run the model locally.

New comment by rapiz in "Superhuman: What can AI do in 30 minutes?"

rapiz — Sun, 26 Mar 2023 11:41:29 +0000

> who are good at leaning into the latest innovations,

maybe not true.

There are many threads where new products got comments of "useless" and then launched. HN users are smart, picky, and not representative for crowds.