Hacker News: sweetheart

New comment by sweetheart in "Humanising LLM Outputs Is Dumb"

sweetheart — Mon, 10 Aug 2026 20:27:55 +0000

This has recently become a pretty pressing issue for me, as it's starting to severely hinder my ability to be productive with the models. It's hard to tell if its getting worse with every model release, specific to Anthropic's models, a reflection of my ADHD, all/none of the above, but holy shit do I get aggravated when I'm forced to parse the most unintelligible, jargon-dense bullshit explanations in whatever the model output is. And then I feel silly getting genuinely tilted by the model's inability to just... explain something semi-normally, without it requiring me to berate it into simplicity.

For some discrete skills I use, I include a final step on the the output that runs through 1+ subagents to de-slop the text and to actually simplify it, but so far nothing has worked as well I've hoped. Considering hopping off Anthropic's models to try out others to see if they're less egregious.

New comment by sweetheart in "Nvidia, Microsoft, Meta warn against overregulating open-weight models"

sweetheart — Fri, 24 Jul 2026 14:51:42 +0000

It makes a ton of sense to see this sort of joint effort being made by these enormous companies, but man, there's some part of me that can't help but feel grossed out knowing the best way to get the US to make the right choice is to have a handful of elite business people say it'd be bad for them.

New comment by sweetheart in "What if remote working, not AI, is to blame for weak junior hiring?"

sweetheart — Mon, 01 Jun 2026 12:34:12 +0000

I’m kind of doing this right now for the first time as a founding engineer in a very small but tight knit team. On paper, I’m objectively in a less secure position, but holy shit is it nice to enjoy work consistently.

New comment by sweetheart in "Steve Wozniak cheered after telling students they have AI – actual intelligence"

sweetheart — Fri, 22 May 2026 11:37:06 +0000

Okay!

New comment by sweetheart in "Steve Wozniak cheered after telling students they have AI – actual intelligence"

sweetheart — Fri, 22 May 2026 11:04:58 +0000

I think it’s possible.

New comment by sweetheart in "Steve Wozniak cheered after telling students they have AI – actual intelligence"

sweetheart — Fri, 22 May 2026 10:54:47 +0000

They can learn the skills to advance research and fill the roles that help determine what sorts of guard rails there should/could be to ensure it’s used in as helpful a manner as possible.

New comment by sweetheart in "Throwing AI-generated walls of text into conversations"

sweetheart — Thu, 21 May 2026 16:01:38 +0000

But the point OP is making is that it's entirely possible that the person doing this _does_ see it as them being as helpful as possible. That doesn't mean it doesn't suck, or that it isn't annoying, though. I dunno, just seems like a coin toss to me: was this backed by good intentions or not? Without other "evidence", assuming that it was well-meaning but misguided feels better for _both_ of us (at least in my experience).

New comment by sweetheart in "We've made the world too complicated"

sweetheart — Sat, 16 May 2026 11:10:38 +0000

I felt this way VERY strongly last year and into the beginning of this year. I was definitely burned out, but that doesn’t mean I wasn’t right in noticing a lot of the same stuff described in the blog post. I was dangerously close to trying to talk my wife into liquidating our 401ks to buy an off grid cabin and resign from modern life.

What helped in the end was seemingly some sort of combination of acceptance + commitment, plus a looot of reflection on the nature of mind/mindfulness. Basically, understanding that our planet is a roiling ball of material simply unfolding over billions of years, and any apparent boundaries between “me” and “everything else” (including all the stressful stuff!) is an illusion caused by my silly limited human capacity to understand and perceive.

Sounds woo-woo and silly, but it has changed my life and provided me a framework to hold both “modern society is a chaotic train wreck” and “the only thing to do is be present and kind” at the same time in a way that’s free of contradiction and completely obvious in hindsight. I hope you feel better soon, blog post author! you deserve to.

New comment by sweetheart in "The Thinking Plant's Man (2025)"

sweetheart — Wed, 06 May 2026 15:41:20 +0000

So, so much of human history is just us trying and failing to find reasons we're unique, as if uniqueness is somehow the source of value, meaning, or joy in life. We really seem to react strongly against the widening of our consideration for the experience of non-human life (or human life, if it looks just a little different).

New comment by sweetheart in "AI's biggest critic has lost the plot"

sweetheart — Tue, 28 Apr 2026 15:41:02 +0000

I'm pleasantly surprised to see this! Last year a few people I know in person, and a podcast I enjoy, talked about or to Ed Zitron and I felt like I was going crazy because so, so much of what he argued was either woefully outdated, or just a fallacy. It's also annoying because it'd be such an interesting topic to explore rigorously and without motive. As mentioned in the article, those analyses _can_ be found. But man, Ed Zitron just seems loud and silly.

New comment by sweetheart in "MeshCore development team splits over trademark dispute and AI-generated code"

sweetheart — Thu, 23 Apr 2026 18:14:09 +0000

yeah

New comment by sweetheart in "Claude Design"

sweetheart — Fri, 17 Apr 2026 17:27:48 +0000

I'm in the same boat. Tailwind always seemed insane to me, even after really giving it the benefit of the doubt and trying it out. I use it now only because its so easy for the LLMs to use, so I don't need to actually interface with it at all.

New comment by sweetheart in "Artemis II is competency porn"

sweetheart — Sat, 11 Apr 2026 11:01:19 +0000

I didn’t follow the mission much as it occurred, but it’s striking to me how much I understand what the author means. Feels like the first event in many, many years that doesn’t amplify the feeling of being in the absurdist nightmare timeline. Artemis II felt like a 2013 event, not a 2026 event!

New comment by sweetheart in "LinkedIn is illegally searching your computer"

sweetheart — Thu, 02 Apr 2026 15:05:36 +0000

the drone that gives hugs, right??? right????

New comment by sweetheart in "Grief and the AI split"

sweetheart — Fri, 13 Mar 2026 11:42:16 +0000

can i come

New comment by sweetheart in "Nvidia PersonaPlex 7B on Apple Silicon: Full-Duplex Speech-to-Speech in Swift"

sweetheart — Thu, 05 Mar 2026 15:26:12 +0000

holy shit I cannot believe how polished that is.

New comment by sweetheart in "I guess I kinda get why people hate AI"

sweetheart — Mon, 16 Feb 2026 18:44:44 +0000

The recent developments of only the last 3 months have been staggering. I think you should challenge your beliefs on this a little bit. I don't say that as an AI fanboy (if those exist), it's just really, really noticeable how much progress has been made in doing more complex SWE work, especially if you just ask the LLM to implement some basic custom harness engineering.

New comment by sweetheart in "I started programming when I was 7. I'm 50 now and the thing I loved has changed"

sweetheart — Tue, 10 Feb 2026 17:16:09 +0000

> I started programming when I was seven because a machine did exactly what I told it to

What a poetic ending. So beautiful! And true, in my experience.

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

sweetheart — Tue, 10 Feb 2026 14:50:34 +0000

This is my first time hearing of Oxide, but I had the same initial thought after reading this blog post then poking through their site. The degree of careful thought put into their policies and culture is really impressive, at least from the outside. Good for them, I hope they continue to be in a position to have that luxury (genuinely).

New comment by sweetheart in "Claude Opus 4.6"

sweetheart — Thu, 05 Feb 2026 22:24:04 +0000

The "UI" is indeed represented in memory in tree-like structure for which positioning is calculated according to a flexbox-like layout algo. React then handles the diffing of this structure, and the terminal UI is updated according to only what has changed by manually overwriting sections of the buffer. The CLI library is called Ink and I forget the name of the flexbox layout algo implementation, but you can read about the internals if you look at the Ink repo.