Hacker News: drakenot

New comment by drakenot in "The AI revolution in math has arrived"

drakenot — Tue, 14 Apr 2026 04:45:24 +0000

I think voice mode uses weaker models, just an FYI relative to the SOTA

New comment by drakenot in "We don't need continual learning for AGI. What top labs are currently doing"

drakenot — Thu, 05 Mar 2026 20:29:18 +0000

It reminds me of the standard counter to the Chinese Room thought experiment: the person inside doesn’t understand Chinese, but the system _does_. The person, the rules, and the lookup tables together form the thing doing the understanding.

New comment by drakenot in "Apple patches decade-old iOS zero-day, possibly exploited by commercial spyware"

drakenot — Thu, 12 Feb 2026 16:27:40 +0000

How can you tell that you were breached?

New comment by drakenot in "Cursor switches pay-per-token when your plan limit end. Calls "On-Demand usage""

drakenot — Wed, 11 Feb 2026 06:58:43 +0000

Claude Code has gotta be way cheaper for this, right?

New comment by drakenot in "Oxide raises $200M Series C"

drakenot — Tue, 10 Feb 2026 14:57:36 +0000

I think its the end-to-end, integrated nature of it.

API driven, have "elastic resources", etc, etc. Rather than bolting together various solutions you get to have a "Cloud-like" stack in your own datacenter.

New comment by drakenot in "Voxtral Transcribe 2"

drakenot — Thu, 05 Feb 2026 13:27:48 +0000

Yes. That is it

New comment by drakenot in "Voxtral Transcribe 2"

drakenot — Thu, 05 Feb 2026 01:19:24 +0000

This past month Parakeet v3 dropped with a streaming ASR model that is 0.6B params, can run on a CPU and is super good.

New comment by drakenot in "Nvidia shares are down after report that its OpenAI investment stalled"

drakenot — Tue, 03 Feb 2026 13:41:42 +0000

I agree that DeepSeek and others have been doing most of the published research for future frontier models.

But I’m not sure we can say they the big US companies aren’t doing foundational research anymore. I suspect a lot of it is just kept unpublished as a result of the intense competitive pressure, which is disappointing.

New comment by drakenot in "Prism"

drakenot — Wed, 28 Jan 2026 17:22:18 +0000

This is handy for maintaining a resume!

I converted my resume to LaTeX with Claude Code recently. Being able to iterate on this code-form of my document is so much nicer than fighting the formatting with in Word/Google Docs.

I dropped my .tex file into Prism and it makes it nice to instantly render it.

New comment by drakenot in "Anthropic Explicitly Blocking OpenCode"

drakenot — Thu, 15 Jan 2026 01:00:04 +0000

Perhaps concentrated use of Claude Code increases their perceived market value.

It also perhaps tries to preserve some moat around their product/service.

New comment by drakenot in "Indifference is a power (2015)"

drakenot — Tue, 13 Jan 2026 14:42:45 +0000

I think that’s more a critique of the modern caricature of stoicism than of Stoicism itself. Classical Stoicism isn’t about suppressing emotions. It’s about understanding your emotions, examining where they come from, and choosing how you respond rather than being ruled by them.

New comment by drakenot in "Dbos: Durable Workflow Orchestration with Go and PostgreSQL"

drakenot — Fri, 03 Oct 2025 03:29:06 +0000

I read the Dbos vs Temporal thing, but can you speak more about if there is a different in durability guarantees?

New comment by drakenot in "Measuring the environmental impact of AI inference"

drakenot — Sat, 23 Aug 2025 04:54:09 +0000

This from quantizing their Gemini model?

There are a lot of anecdotal reports of quality differences following some Gemini 2.5 Pro releases earlier in the year.

New comment by drakenot in "Kodak says it might have to cease operations [updated]"

drakenot — Thu, 14 Aug 2025 02:12:20 +0000

> Presteign did not care for the artists, musicians, and fops Olivia kept about her, but he was pleased to see a scattering of society notables this morning.

> There was a Sears-Roebuck, a Gillet, young Sidney Kodak who would one day be Kodak of Kodak, a Houbigant, Buick of Buick, and R. H. Macy XVI, head of the powerful Saks-Gimbel clan.

- The Stars My Destination

New comment by drakenot in "GPT-5: Overdue, overhyped and underwhelming. And that's not the worst of it"

drakenot — Sun, 10 Aug 2025 11:46:13 +0000

Hilarious. It’s impossible to have an original thought!

New comment by drakenot in "GPT-5: Overdue, overhyped and underwhelming. And that's not the worst of it"

drakenot — Sun, 10 Aug 2025 00:42:55 +0000

Fixed, thanks.

New comment by drakenot in "GPT-5: Overdue, overhyped and underwhelming. And that's not the worst of it"

drakenot — Sun, 10 Aug 2025 00:38:04 +0000

Gary Marcus tends to have pretty shallow analysis or points.

His takes often remind me of Jim Cramer’s stock analysis — to the point I’d be willing to bet on the side of a “reverse Gary Marcus”.

New comment by drakenot in "Yet Another LLM Rant"

drakenot — Sat, 09 Aug 2025 15:33:59 +0000

ChatGPT 5> Without adding third-party dependencies, how can I compress a Data stream with zstd in Swift on an iPhone?

(It auto-routed to its thinking model, took 25s)

> Short answer: you can’t. iOS doesn’t ship a public zstd encoder/decoder. The system Compression and AppleArchive frameworks only expose lzfse, lz4, zlib/deflate, and lzma. So with only Apple SDKs, there’s no way to produce zstd-compressed bytes.

> > Two practical paths:

> 1. If zstd isn’t mandatory, use a built-in codec (lzfse or lz4 are fastest) via Compression’s streaming API. Example (pure Swift, no deps) that compresses an arbitrary byte stream: >

> 2. If zstd is mandatory, you must embed zstd yourself. That can mean:

> - Vendoring the official C sources (libzstd) directly into your project and calling them via a tiny Swift wrapper/bridging header

> - Statically linking a prebuilt libzstd.a you compile for iOS



New comment by drakenot in "OpenAI's ChatGPT Agent casually clicks through "I am not a robot" verification"
drakenot — Thu, 31 Jul 2025 12:40:38 +0000

This is why this stuff is going to shift to the user’s AI enabled browser.
Half of the sites already block OpemAI.  But if it is steering the user’s browser itself?



New comment by drakenot in "GLM-4.5: Reasoning, Coding, and Agentic Abililties"
drakenot — Mon, 28 Jul 2025 21:21:43 +0000

The commentary around every Chinese model is incredibly disappointing. Asking about Tiananmen Square isn't some clever insight.
Look at the political leanings that government-backed AIs in the United States will soon be required to reflect: those of the current administration.
I was hoping to hear from people reporting on their utility or coding capabilities instead.