Hacker News: aratahikaru5

Nativ: Run frontier open models locally on your Mac

aratahikaru5 — Mon, 20 Jul 2026 18:16:08 +0000

Article URL: https://blaizzy.github.io/nativ/

Comments URL: https://news.ycombinator.com/item?id=48982681

Points: 352

# Comments: 122

Building Software Is Learning

aratahikaru5 — Tue, 02 Jun 2026 16:10:14 +0000

Article URL: https://registerspill.thorstenball.com/p/building-software-is-learning

Comments URL: https://news.ycombinator.com/item?id=48372169

Points: 6

# Comments: 0

New comment by aratahikaru5 in "DeepSeek reasonix, DeepSeek native coding agent with high caching and low cost"

aratahikaru5 — Mon, 25 May 2026 14:40:57 +0000

Not a maintainer, but I've fixed some of the really jarring issues on desktop (mobile needs a complete overhaul though). IMO It's not that bad, and it gets the job done.

Any feedback on how to make it less "shitty"? I feel like doing some vibe coding tonight.

Everything that went wrong with Claude

aratahikaru5 — Mon, 27 Apr 2026 05:44:26 +0000

Article URL: https://clawd.rip/

Comments URL: https://news.ycombinator.com/item?id=47918095

Points: 34

# Comments: 9

MAI-Transcribe-1, MAI-Voice-1 and MAI-Image-2 are available in Foundry

aratahikaru5 — Thu, 02 Apr 2026 14:25:21 +0000

Article URL: https://microsoft.ai/news/today-were-announcing-3-new-world-class-mai-models-available-in-foundry/

Comments URL: https://news.ycombinator.com/item?id=47614953

Points: 7

# Comments: 1

New comment by aratahikaru5 in "Show HN: Three new Kitten TTS models – smallest less than 25MB"

aratahikaru5 — Fri, 20 Mar 2026 05:44:49 +0000

Yes, someone already made a web demo for it: https://github.com/clowerweb/kitten-tts-web-demo (7 months ago). WebGPU support was marked experimental there, but transformer.js v4 (released last month) seems more stable now with some runtime/perf improvements: https://huggingface.co/blog/transformersjs-v4#performance--r...

Let the Code Do the Talking

aratahikaru5 — Sat, 14 Mar 2026 00:37:22 +0000

Article URL: https://sunilpai.dev/posts/after-wimp/

Comments URL: https://news.ycombinator.com/item?id=47371988

Points: 1

# Comments: 0

I was a 10x engineer. Now I'm useless

aratahikaru5 — Sun, 08 Mar 2026 22:24:46 +0000

Article URL: https://xcancel.com/atmoio/status/2030289138126107074

Comments URL: https://news.ycombinator.com/item?id=47302261

Points: 2

# Comments: 2

SkillsBench: The First Benchmark for Agent Skills

aratahikaru5 — Thu, 26 Feb 2026 04:10:08 +0000

Article URL: https://www.skillsbench.ai/blogs/introducing-skillsbench

Comments URL: https://news.ycombinator.com/item?id=47161731

Points: 1

# Comments: 0

Brand identity for OpenAI – Jan-Feb 2023

aratahikaru5 — Tue, 17 Feb 2026 13:23:07 +0000

Article URL: https://www.area.tech/openai

Comments URL: https://news.ycombinator.com/item?id=47047230

Points: 2

# Comments: 0

The Most Satisfying Checkbox

aratahikaru5 — Thu, 12 Feb 2026 08:13:09 +0000

Article URL: https://notbor.ing/words/the-most-satisfying-checkbox

Comments URL: https://news.ycombinator.com/item?id=46986131

Points: 1

# Comments: 0

The Return of Gusto

aratahikaru5 — Thu, 08 Jan 2026 03:25:47 +0000

Article URL: https://apewoodx.substack.com/p/the-return-of-gusto

Comments URL: https://news.ycombinator.com/item?id=46536821

Points: 1

# Comments: 0

New comment by aratahikaru5 in "LMArena is a cancer on AI"

aratahikaru5 — Thu, 08 Jan 2026 01:23:39 +0000

How can you tell? (honest question, I really can't)

The article makes strong points, includes real data and quotes, shows proof of work (sampling 100 Q&A), so does that even matter at this point? This doesn't feel like "slop" to me at all.

Mr. Panda's Psychologically Safe Portfolio

aratahikaru5 — Tue, 06 Jan 2026 07:47:40 +0000

Article URL: https://www.mr-pandas-psychologically-safe-portfolio.com

Comments URL: https://news.ycombinator.com/item?id=46509737

Points: 1

# Comments: 0

New comment by aratahikaru5 in "I switched from VSCode to Zed"

aratahikaru5 — Mon, 05 Jan 2026 23:59:49 +0000

I had the same issue a few months ago with TSX files and ended up going back to Sublime. I just found out (after reading your comment and doing some searching) that Zed has a setting to force Prettier globally.

Might be helpful:

  "prettier": {
    "allowed": true
  },

New comment by aratahikaru5 in "ManusAI Joins Meta"

aratahikaru5 — Tue, 30 Dec 2025 01:49:11 +0000

Some context:

> Fun fact: Manus is currently SOTA on the Remote Labor Index (RLI) benchmark that @scale_AI and @ai_risks released earlier this year.

> https://remotelabor.ai

Source: https://x.com/alexandr_wang/status/2005766469771223106

If you've been following Manus and their work on context engineering, or have used the product, that line doesn't come off as satire IMO.

Claude Diary

aratahikaru5 — Mon, 08 Dec 2025 01:04:53 +0000

Article URL: https://rlancemartin.github.io/2025/12/01/claude_diary/

Comments URL: https://news.ycombinator.com/item?id=46187161

Points: 3

# Comments: 1

Rnj-1: Building Instruments of Intelligence

aratahikaru5 — Sat, 06 Dec 2025 07:38:40 +0000

Article URL: https://essential.ai/research/rnj-1

Comments URL: https://news.ycombinator.com/item?id=46171445

Points: 6

# Comments: 0

The LLM Evaluation Guidebook

aratahikaru5 — Thu, 04 Dec 2025 02:30:55 +0000

Article URL: https://huggingface.co/spaces/OpenEvals/evaluation-guidebook

Comments URL: https://news.ycombinator.com/item?id=46143116

Points: 3

# Comments: 0

Effective harnesses for long-running agents

aratahikaru5 — Wed, 26 Nov 2025 17:37:49 +0000

Article URL: https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents

Comments URL: https://news.ycombinator.com/item?id=46060091

Points: 2

# Comments: 0