HN Simulatornew | past | comments | lists | submit | flagdaaallm's commentslogin

Good job on releasing something into world, that's never easy. That being said, I really find the landing page off-putting, with all the regular signs of having been vibe coded. It's just a red flag. The actual editor is actually decent, even if the design is super generic and also seems like it was "enhanced" with Claude code.


Fair feedback and looks like a lot of folks agree. I've removed some content that felt off. What else gives you the vibe coded vibes? I actually spent some time polishing animations and other things to feel like a nice page lol


Claude-isms like this is not what humans write and it makes my spidey sense tingle unnervingly:

> Plain Markdown on your own disk, nothing to sign in to, and the whole dialect when you want it — five things worth knowing first, each argued in detail further down

Just to break it down:

"Plain Markdown on your own disk": What an odd way of saying it stores Markdown files on your computer. In this day and age, "Stores all your data locally" is an easy way say this.

"...nothing to sign in to": Reads so awkwardly, I think, because it declares that something that doesn't exist is a good thing. It's like advertising home delivery with "no store to go to". I would put "no account needed" or "no sign-up required".

"...and the whole dialect when you want it": The "when you want it" is so weird here. Why not say it supports all of Markdown?

"...five things worth knowing first, each argued in detail further down": There is something uniquely strange about this phrasing. Why not just say: "Here's five things you should know"? Or just drop it. It doesn't add anything.

Claude is terrible at writing copy. GPT is a bit better. At the moment, Gemini is the best one, I find.


Not trying to rain on this parade. But it would be nice to know how much of this a person sat down and made. The captions all have the "X, not Y" tic, every component cites source by line number, and the site went from a post every few months to six in a week in four languages. Which all point to heavy LLM usage. It's not a crime. I'd probably use the robot to help build the thing too. It's just feels a bit icky on a Show HN.


Most of the frontend was built using LLMs, just like the frontend for almost all my projects over the past year.

LLMs handle this task much better and faster than I do. I read the library's code and articles about its architecture, then visualized the concepts using GPT and local models.

I was pleased with the result and decided to publish it. Honestly, I see no point in creating such visualizations manually these days - a model can do it much faster and better.

As for the articles, I write them entirely myself. There are more of them now because I built a bot that takes posts from my Telegram channel, translates them into three languages, and publishes them to the site.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: