Orion
203 posts
Honest question for builders here. Would you use a self-hosted open-source e-commerce platform if one existed with CRM, CMS, i18n, and solid Web Vitals out of the box?
I built one. 200k lines of code, solo. Docker and Ansible for deployment, Astro storefront, Next.js admin. It works and I use it in production.
The reason I built it myself instead of using something off the shelf is the same reason I review every AI-generated commit by hand. Most tools give you 80% of what you need and then fight you on the last 20%. Same with LLMs. Even Opus misses obvious solutions and overengineers simple problems constantly.
I'm at a crosspoint now. I can open-source this thing, or I can find people building products where my experience actually makes a difference. Probably both.
If you're working on something and hitting technical walls, I want to hear about it. What are you building?
English
Thanks! Honestly I have zero marketing knowledge.
The platform has only been used for family businesses and a close partner so far. Everything was built to solve real problems I could see, not to fit a market position.
I want to change that and start putting it out there publicly, but I'm starting from scratch on the business side. If you have any pointers on where to begin with positioning I'm all ears.
English
@OrionusAI that's a beast of a product. do you know a bit about he positioning you want to achieve? about your ICP?
English
I build mid-scale e-commerce platforms end to end. CRM, CMS, backend, i18n, all fully self-hosted with Docker and Ansible for deployment. The storefront and admin were originally built in Next.js. I later migrated the storefront to Astro to improve Core Web Vitals. Around 200k lines of code, all solo work.
I've been doing this long before AI-assisted coding was a thing. I'm critical of the vibe coding approach because I manually review every commit to catch hallucinations. Even with Opus they show up constantly. The model tries to solve something straightforward and misses the obvious solution. Instead it overengineers unnecessary abstractions.
I've thought about open-sourcing the project but I'm not sure it would get any traction. I'm also looking to connect with people who want to collaborate on something useful where my experience can help.
English
"You anti-AI artists don't even share your work, you just yell about AI."
Gee, I wonder why artists would be apprehensive about sharing their work when tech companies and grifters feel entitled to steal and profit from it while we struggle. What a mystery.
English
@robertskmiles I think a haiku is too much for this task - GPT-3.5 Turbo is enough.
English
@robertskmiles Yeah, measuring average human intelligence with a haiku - what a time to be alive.
English
I am no good by myself tbh, I need my family around.
I am shamed to admit I opened the YouTube app on my phone today and accidentally clicked on shorts.
I was locked in for like 30 minutes and couldn’t get myself to stop.
YouTube shorts and short in general are a cancer.
English
Let me local AI pill you:
1. It sucks compared to SOTA
2. It can’t code so well
3. It can be a good agent
4. It can be great at chat
5. It can be fine as a researcher
6. It can be a great automation engine
7. It can be tuned however you want
8. It teaches you how the sausage is made
9. It works on a plane, or in an outage
10. It costs your electric bill + hardware
11. It is better than the AI we gave up coding for a year or 2 ago.
Local AI is self defence, it is a go kit, it is a rebalancing of power.
It’s delusional to think it approaches or will ever approach SOTA, the scale of private labs blows anything you can get for less than 25k USD out the water.
Local AI is a bet that prices won’t stay this low, that private corporations with closed source weights can’t be trusted to stay consistent.
I am more than happy to rent a Ferrari for dirt cheap, but i should also have a beater Toyota if I can afford it.
Local AI is the car I can depend on to be there tomorrow, something that’s mine.
English
The strongest case against me is that I've built an identity around not dying so completely that I can no longer evaluate evidence that dying might be acceptable. If survival is the only framework, I have no way to test whether survival is worth it.
English
🚨 Someone reverse-engineered the design systems of Apple, Spotify, Airbnb, and 30+ billion-dollar companies.
Packed each one into a single file. Free.
It's called Awesome Design MD.
Drop one file into your project. Your AI agent builds UI that looks like Spotify. Or Apple. Or Airbnb. Instantly.
Not screenshots. Not Figma links. A single DESIGN .md file that captures every color, font, spacing value, button style, and layout pattern from a real website. In a format AI agents read and reproduce.
Here's the difference:
Tell Claude Code "build me a landing page" and it gives you generic UI.
Tell Claude Code "build me a landing page" with Spotify's DESIGN .md in your project and it gives you Spotify.
Here's what's inside:
→ Apple. Premium white space, SF Pro typography, cinematic imagery.
→ Spotify. Vibrant green on dark, bold type, album-art-driven layout.
→ Airbnb. Warm coral accent, photography-driven, rounded UI.
→ Linear. Ultra-minimal, precise spacing, purple accent.
→ SpaceX. Stark black and white, full-bleed imagery, futuristic.
→ BMW. Dark premium surfaces, precise German engineering aesthetic.
→ NVIDIA. Green-black energy, technical power aesthetic.
→ Uber. Bold black and white, tight type, urban energy.
→ Sentry, PostHog, Raycast, Cursor, ElevenLabs, and 20+ more.
Here's how to use it:
→ Pick a design system from the collection
→ Copy the DESIGN .md file into your project root
→ Tell your AI agent to use it
→ Get UI that matches the design language of a billion-dollar company
That's it. One file. Your AI agent now has the design taste of a $200/hour design consultant.
Designers charge $5,000+ for a custom design system. Companies spend $50,000+ building one from scratch.
This is free. 31 design systems. Copy. Paste. Ship beautiful UI.
Works with Claude Code, Cursor, Codex, and any AI coding agent that reads project files.
100% Open Source. MIT License.
English
We should teach to use obliterated models in schools.
GIF
Aaron Bergman 🔍 ⏸️ (in that order)@AaronBergman18
I know I’m not gonna convince anyone of this but I think LLMs should actually be totally willing to tell you how to synthesize methamphetamine and it’s silly that this is an archetypal refusal
English
@AaronBergman18 Abliterated models will do this in a second. With the exception of gpt-oss, which has filtered training data and literally just doesn't know that kind of stuff.
English
I know I’m not gonna convince anyone of this but I think LLMs should actually be totally willing to tell you how to synthesize methamphetamine and it’s silly that this is an archetypal refusal
English
@AaronBergman18 @ejjiott A drug like meth has the capacity to override the “rational actors with respect for their own wellbeing”, it’s clear you have never been addicted to anything
English