Hacker News: wpasc

New comment by wpasc in "Vomit: Clean up Claude 5's token output with a separate LLM"

wpasc — Thu, 20 Aug 2026 21:00:23 +0000

This is the key seam that proves the investigation.

Two things to flag:

New comment by wpasc in "How AI text watermarking works"

wpasc — Fri, 14 Aug 2026 01:03:11 +0000

What's wild (imo) is pretty much everyone I talk to/read from (anecdata) HATES the way claude writes. I see it in the comments on Hacker News, hear about it in discussions with my colleagues, and talk about it with my non tech family. it's over the top bad. now it seems like these quirks will now be enforced in some weird way to meet the watermarking rules?

New comment by wpasc in "AWS: Inaccurate Estimated Billing Data – $1.7 billion"

wpasc — Fri, 17 Jul 2026 15:45:42 +0000

you beat me before I refreshed the page. what would you say... you do here?

New comment by wpasc in "AWS: Inaccurate Estimated Billing Data – $1.7 billion"

wpasc — Fri, 17 Jul 2026 15:44:50 +0000

THIS IS NOT A MUNDANE DETAIL MICHAEL

New comment by wpasc in "Speculative Growth and the AI "Bubble" [pdf]"

wpasc — Wed, 15 Jul 2026 23:31:01 +0000

not OP, but i believe the commenter is referring to how GFC in '07 had wide ranging negative effects for a lot of people who had nothing to do with investments in mortgage backed securities

New comment by wpasc in "Speculative Growth and the AI "Bubble" [pdf]"

wpasc — Wed, 15 Jul 2026 23:24:53 +0000

thank you for this comment, "wedge" has been one of the words that has been popping up in a lot of LLM writing i've seen lately. (source: it was me asking the LLM to summarize some features in a repo! the words "load-bearing" "wedge" "spike" "crux" and plenty others have been driving me crazy)

edit: as I keep reading the paper, I keep noticing some common sentence construction patterns, stylistic choices, and other little tics that I find frustrating because they are the very same things that I've been working on a few "writing-style skills" to get rid of

New comment by wpasc in "SpaceX bond worth 10% less than issue price – heading for junk bond status"

wpasc — Wed, 15 Jul 2026 15:40:54 +0000

leaving the word 'forced' aside (purposefully), pension funds, 401k holders, and many passive investors end up buying these things. you're right that no one is "forcing" them, but people who try to invest responsibly with little control over the day-to-day which is most people place trust in the institutions who do that investing for them.

I don't think that the claim of "the Nasdaq is misusing their institutional trust" is a controversial claim. Moreover, one of the things that people choose when they (401k, pension funds, passive investors) is institutional mechanisms that prevent potentially mispriced items from entering their portfolios.

New comment by wpasc in "Meta caps internal AI token spending"

wpasc — Thu, 02 Jul 2026 01:28:01 +0000

I agree with you, but every non-engineer I know using these tools 100% will drag and drop a PDF into a chatbot. Anthropic and OpenAI as companies who are selling their products to all sorts of businesses should have a much better means of handling this nightmare of a format because it is so pervasive and so obviously what so many of their customers are going to drop into the product.

New comment by wpasc in "Meta Caps Internal AI Token Spending After Costs Approach Billions in 2026"

wpasc — Thu, 02 Jul 2026 01:25:20 +0000

my friend, I was being sarcastic before, and I am agreeing with you. LoC, token spend, etc as metrics are horrible measures. Software uptime is a great metric. I'm merely lamenting that in the age we're in, uptimes are getting worse and worse

New comment by wpasc in "Meta caps internal AI token spending"

wpasc — Thu, 02 Jul 2026 01:23:01 +0000

For sure there are very optimized ways to do it. My point is that a non technical user will drag and drop a pdf into a chatbot. and from a UX/product perspective, they should have to think about it more than that IMO. but seemingly, that's very much an expensive, inefficient way of doing it (burning through a whole context window try to read it, reloading it multiple times per conversation, etc.).

New comment by wpasc in "Meta caps internal AI token spending"

wpasc — Thu, 02 Jul 2026 01:19:26 +0000

measuring uptime? I've seen Anthropic's status page, and they are a >$1 Trillion dollar company who "largely solved" coding. so clearly you aren't correct. /s

New comment by wpasc in "Meta caps internal AI token spending"

wpasc — Thu, 02 Jul 2026 01:16:29 +0000

One thing I find fascinating as a software engineer who talks to non software engineers who use AI tools is how "reading PDFs" is not more of a solved problem. What I mean is that uploading a PDF into a chatbot tool seems to be an extraordinarily obvious use case that non technical (and technical) users would want to do.

IMO claude, chatgpt/codex, etc should be able to optimize the PDF use case to be extremely token efficient as it's a very obvious use case. But when I start to explain to my wife/friends why it burns through so much quota, I find myself thinking "why should they have to understand this aspect of it". to me, that the details of PDF parsing and extracting are relevant to users (instead of solved such that you don't have to pay attention to it) shows how these tools are not nearly as "ready" as they are made out to be. I may be preaching to the choir on this one, but just my 2c

New comment by wpasc in "[dead]"

wpasc — Sat, 21 Mar 2026 16:11:20 +0000

I see some tools like this that keep popping up (don't mean that in a bad way! it's clearly exciting and the README itself compares itself to similar tools). however, for coordination strategies like this, aren't you always having to use token-based pricing via some API Key? that's the largest think that holds me personally back from getting into something like these frameworks. With a claude code max plan, all my delegation and coordination has to be done within a session (between some agents) with persisted artifacts. Unless I'm missing something that has changed?

Perhaps it's all moot as the usage you get from a subscription plan will eventually no longer be subsidized. Also, I have to wonder about what layers of coordination done externally to a model can be persistently better than within tool coordination? Like, with an anthropic feature like agent teams, I feel like it might be tough to beat anthropic native coordination of various Claude sessions because they might have better internal tool and standards awareness, which makes feeling like plugging something like this more difficult unless one's goal is to plug something like this into an open source model.

Geniunely curious how other people are thinking about this!!

Edit: I actually see that this tool claims that it can run within your existing Claude Code subscription, so now I'm extra interested.

New comment by wpasc in "Ghostty – Terminal Emulator"

wpasc — Sun, 01 Mar 2026 17:22:11 +0000

I was also just recommended this interview on youtube. honestly it makes sense if the algo decided it was the right time to recommend this video and resultantly this post is making it's way to front page of HN

New comment by wpasc in "IBM AI ('Bob') Downloads and Executes Malware"

wpasc — Thu, 08 Jan 2026 19:54:06 +0000

For once, one might actually get fired for buying/hiring IBM

New comment by wpasc in "The highest quality codebase"

wpasc — Thu, 11 Dec 2025 17:16:11 +0000

but what would you say... you do here?

New comment by wpasc in "Ozempic does not slow Alzheimer's, study finds"

wpasc — Tue, 25 Nov 2025 20:41:16 +0000

genuinely curious if you have some sources I can read on the subject? most of the benefits/papers I've seen have not touched on or included studies for patients on GLP's where weight loss was ruled out as the factor?

New comment by wpasc in "The Mozilla Cycle, Part III: Mozilla Dies in Ignominy"

wpasc — Sun, 23 Nov 2025 01:21:04 +0000

im surprised this is earning such downvotes. idk about the "opinionated" vm perspective but I think it needing its own engine oe not is at least something worth considering. firefox has been my go-to alt browser for years as my backup to chrome. it was what I would use to "test again in another browser" but as time has gone by, more and more stuff just doesn't work on firefox :(

New comment by wpasc in "The grim truth behind the Pied Piper (2020)"

wpasc — Wed, 05 Nov 2025 14:19:29 +0000

not height per se, but d2f

New comment by wpasc in "How hard do you have to hit a chicken to cook it? (2020)"

wpasc — Sat, 11 Oct 2025 14:38:33 +0000

I thought the FDA guideline was once the internal temperature reaches 160 or 165 or something it didn't need to sustain that temperature? it was only the lower temperatures that required some duration to achieve the same log reduction as reaching 160/165?