Hacker News: lebuin

New comment by lebuin in "A new spam policy for “back button hijacking”"

lebuin — Tue, 14 Apr 2026 11:19:11 +0000

You can also click the address bar and then press you shortcut. Should be faster and works for all shortcuts AFAIK.

New comment by lebuin in "So where are all the AI apps?"

lebuin — Tue, 24 Mar 2026 15:12:51 +0000

I'd argue that LLMs are not yet capable of the last step, and because most sufficiently large AI-generated codebase are an unmaintainable mess, it's also very hard for a human developer to take over and go the last mile.

New comment by lebuin in "Death to Scroll Fade"

lebuin — Wed, 18 Mar 2026 18:17:23 +0000

God yes. For some reason, I automatically scroll in such a way that I always keep what I'm reading at the very top of the screen. Which means that every time I want to reread a sentence I first have to scroll past the header.

New comment by lebuin in "Ask HN: What Are You Working On? (March 2026)"

lebuin — Mon, 09 Mar 2026 08:10:58 +0000

I'm building web-based CAD software for woodworkers. Not a plugin, I'm starting from scratch. I'm aiming for it to be intuitive for non-technical users (think SketchUp), while also offering some of the more powerful tools of "proper" CAD tailored for woodworking: simple parametric workflows, cutting layout optimization, built-in tools like chamfers and joints,...

https://maqet.app

New comment by lebuin in "How we rebuilt Next.js with AI in one week"

lebuin — Wed, 25 Feb 2026 11:25:30 +0000

If you are using webpack, see if you can make the switch to turbopack. It cut my build times from ~1 minute to 15 seconds, incremental builds are down from 10 seconds to 2. Memory usage is down a ton as well. But if you rely on webpack plugins this may not be an option for you.

New comment by lebuin in "AI makes you boring"

lebuin — Thu, 19 Feb 2026 23:41:07 +0000

For me, the answer to this question is: parts that involve no architectural decisions, and that won't need to be extended or built upon significantly in the future.

When I'm working on a greenfield project that I intend to build out further (which is what I am currently doing), I find that there's not a lot of work that fits those criteria. I expect that can change drastically when you're working on something that is either more mature, or more narrowly scoped (and thus won't need to be extended too much, meaning poor architectural decisions are not a big issue).

New comment by lebuin in "27-year-old Apple iBooks can connect to Wi-Fi and download official updates"

lebuin — Thu, 19 Feb 2026 08:04:57 +0000

I had to do a fresh install on a 2015 iMac. Same problems with the SSL certificates. I found it rather shocking that a 10 year old computer cannot be booted anymore, and as far as I understand it it's mostly because apple chooses to serve certificates with poor backwards compatibility on a domain that is used for updates, which is just lazy.

New comment by lebuin in "Thinking hard burns almost no calories but destroys your next workout"

lebuin — Tue, 17 Feb 2026 08:35:20 +0000

I think OP may be referring to the idea that the total number of calories burned in a day doesn't meaningfully change under a workout regime. Working out does burn calories, but after a few session your body starts to compensate by burning less calories in other areas (e.g. immune and reproductive system). The net result is close to zero, except in very demanding workout regimes.

I don't have the background to fully evaluate how true that is. I read "Burn" by Herman Pontzer, which at least makes a very good case for it.

New comment by lebuin in "Evaluating AGENTS.md: are they helpful for coding agents?"

lebuin — Tue, 17 Feb 2026 08:08:40 +0000

It seems like LLMs in general still have a very hard time with the concepts of "doubt" and "uncertainty". In the early days this was very visible in the form of hallucinations, but it feels like they fixed that mostly by having better internal fact-checking. The underlying problem of treating assumptions as truth is still there, just hidden better.

New comment by lebuin in "Breaking the spell of vibe coding"

lebuin — Sun, 15 Feb 2026 12:36:48 +0000

Exactly. The fact that an LLM isn't very good at helping you fix basic organizational issues like this is emblematic. Quoting the article: "We have automated coding, but not software engineering."

New comment by lebuin in "My AI Adoption Journey"

lebuin — Fri, 06 Feb 2026 13:01:21 +0000

There's also no canonical way to write software, so in that sense generating code is more similar to coming up with a potato soup recipe than compiling code.