Hacker Newsnew | past | comments | ask | show | jobs | submit | madprops's commentslogin

For the last couple of months I was working daily, and intensely, on Meltdown. The project started in 2024, but I retook it recently and improved/extended it a lot.

https://meltdown.merkoba.com

This is an AI Harness + Platform, made in Python and Tkinter.

It has a lot of advanced features, arguments, commands, shortcuts.

Lots of widgets and the markdown parser were done from scratch.

It has a powerful API to create conversational sessions interactively.

You can learn about it on a blog post I made:

https://github.com/madprops/blog/blob/main/docs/meltdown/mel...


Wow! Let me be the first to crangukate you. It has lots of features and commands and arguments. I've been thinking about this for a long time. I've always said that we need commands and arguments. Maybe a cli. Keep it going bro!


I don't know if you are being funny, but I say stuff like "arguments" because it has 452 of them, because I wanted to make almost everything configurable.


Hah. Just now I was reading about how Sony hated Michael Jackson because he owned half of the music licenses, and some say Sony might be been involved in his demise. My point being, maybe Sony is not a nice company.


Great new way to write blog posts!


I tend to like images in a split second, when they try to achieve a certain aesthetic at least. I can later inspect them more carefully and understand why they're good.

Reading text, it can either grab me or it doesn't. I've asked about that to an AI and there's some reason for that. Some text is more compatible with me, due to the rhythm or combination of chars/words used, how they actually look/feel, it's not just about the content. So I'm assuming some writers are going to be more compatible with me while others will make me dismiss them quickly or just skim.

I like to think I have the correct objective taste in music, but everybody thinks that about themselves.


> later inspect them more carefully and understand why they're good

Reductionism is hard to apply to intuitions, and we often fool ourselves.

I personally call this the obviously test: smart people correctly notice something and say it is obvious. It is a rather rarer skill to understand and explain why. I have particularly noticed it with coaching (which adds the next levels of difficulty in getting someone to grok what you've seen).


Just use both. vscode + copilot for autocomplete and antigravity for prompting. My main editor is vscode and I do heavy manual edits, and I also heavily use antigravity. I feel like I've become very productive and not actually renouncing the ways of the programmer, but augmenting them, as long as I'm careful enough with the generated code by checking every file it touched.


Some months ago I made this Morse Code Universe where people travel through zones, stumble upon each other, and communicate in morse code.

https://mall.merkoba.com/

All they can do is use morse code to communicate, even the names are assigned automatically. There is a zone map with hundreds of zones, zones with recent activity get a red hue. Secret zones can be unlocked if you happen to use one of the sekrit morse code words.

It's a casual mmorpg/townsquare that is fundamentally safe since the best you can do is focus to type a very offensive word in 20 seconds.

I stumbled upon some random people who visited the site as I was developing it. Taught some the words to enter the secret zones as a game, took them on a tour. Also met a morse code aficionado and we had a little conversation in morse code, eventually met him in another site I have.


AI is probably old italian tech


Concerning globo-list. Centralization/takeover, aka an eventual "we will manage you"; which might be the true colors of the Linux Foundation. Forks would just get absorbed and used internally instead of depending on the performance of random informal earth citizens. The site is not even pleasant to read with that font. Villainy is parodied in this world heavily, names like Discord, Palantir, AI Companies talking about doom scenarios and enjoying it: so it's cool and expected to be a villain, to wrestle with the other kinds of power. I just want some fresh choices to polish the kind of company I want to get around me, which would likely be the opposite of who signed that letter.


Felicidades. Sometimes I think about returning to a public university to have something structured to do, especially since I've been in a way educating myself through AI and I appreciate the idea of an environment dedicated to people with a similar intellectual interest. I don't really care about academia, if I do it it would be an experimental change of environment. If there is even a place for people like me in that public university at all (older people).


I know of a pensioner who successfully did up to a Masters in astronomy, so there are definitely examples of older people as students in academia (not to mention many the professors are older, and especially the emeritus professors).


I think it’s worth a try. Being a bit older than the rest (almost thirties) when I started also gave me pause at first. But I felt that people were very welcoming regardless of your situation.


All you need is a gun and a car. AI psychosis if you don't have a heart.

Remember that Terrans are the most based race in StarCraft. Protoss got too stiff.


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

Search: