AI, ML, LLMs, SD, coding, music. https://xitter.gq/t.co/8zwGUMRrcT
Sagarnata
Joined January 2009
- Tweets20.9K
- Following2.9K
- Followers1.2K
- Likes28.6K
Pinned Tweet
newsbubbles.github.io/ismail… Suno and Udio, meet Claude Code and Agents.
Time to step up.
@higgsfield_ai sent us a cease and desist last night, asking us to shut down our Open Source Higgsfield challenge.
Our read on why they sent it. When the apps people pay for can be rebuilt in the open in two weeks, that’s an uncomfortable thing for a closed platform to watch.
So here’s us reiterating our promise to every builder in the challenge and community, and we’re keeping it:
• The challenge stays open
• Final submissions close tomorrow, Oct 30
• Winners are announced Nov 6, with prizes paid as promised
You showed up for each other. We’re showing up for you.
If we win, we all win!
Comfy out 🫡
Calling any ornithologists or bird watchers. I have a blind listening test for bird calls. Can you tell which one is real? claude.ai/artifact/QzGojzEPU… ... you can copy your answers back at the end and DM me.
Nate Codes retweeted
This is what we made on last nights stream over the course of several hours.
36 Chambers: Preying Mantis vs Drunken Sloth
A lot of fun times were had and cool things learned.
Appreciate everyone who hung out with me!
MidJourney Niji + Seedance 2.0 + Night Shift
Finally someone said it.
what if the solution to the cyber apocalypse is to roll your own software?
- we use human slopware libraries because code used to be expensive
- turns out these libraries have tons of vulnerabilities
- exploiting them makes economic sense as they’re deployed at scale
—> what if we just don’t use libraries and have agents write idiosyncratic custom solutions/versions from scratch?
those still will have vulnerabilities but the economics of attack change significantly?
the whole npm this npm that is from an era where software was hard and expensive, no?
Nate Codes retweeted
The original report was 61 pages, it was redacted to 38.
Nate Codes retweeted
The AI world is built for devs. The real world is run by flower shop owners.
Almost everything in AI right now assumes you use GitHub, live in a terminal, and know how to write a decent prompt. Someone selling flowers or fixing roofs has zero time to learn any of that.
For the next 30 days, I’m building the bridge, as part of 30 Days of Boardy.
The idea is “simple”: a local-first desktop app. You install it, it checks your hardware, picks a model that’ll actually run on your machine, and explains everything in plain human words. No tokens, no inference, no context windows.
Everything else comes through a marketplace of vetted plugins, named after the problems they solve, not the tech behind them. As little learning curve as I can get away with.
Where I’m at on Day1:
- hardware detection semi-works (as far as I can test it on my own machines)
- Ollama baked in
- basic local chat
- marketplace shell
- one plugin that doesn’t do squat :]
Where I want to be on day 30: a working platform with 2-3 plugins a non-technical owner can actually use. Packaged, wrapped, and installable by anyone who wants it.
I also have a full-time job, so... let’s see how this goes ;)
#BuildingInPublic #boardyai
