I've had friends homeless recently, Costco was the basis of the best choices of their poverty finance even when living out of a car.
I think it's accessible to even the poorest people who work in the US, but it doesn't mean it's cheap for them or worthwhile without a home/reasonable commute.
Time is a major commodity for people working 2 or more jobs and an hour and back commute to Costco is often not worth it.
What I don't understand is how half the comments are calling out how bad the content is, yet it's somehow 4th on the frontpage?
It looks like generic AI slop, the site doesn't even render the headings for their SEO spam "Curated AI Tool Collections by Use Case" section properly and they're half cut off. The images all have the very distinct generic AI hue to them without any attempt of bringing it into a specific style or brand.
Who is upvoting this stuff? Do people not care? Is it just bots gaming the system? Am I an old man shouting at a cloud?
Some of these acquired products aren't dead or even in maintenance mode though, they're still running. So it doesn't really make sense to include them in a graveyard, among the other very much alive projects like Streamlit and LangFuse, or missing dead projects like Sora.
This is just slop. I wouldn't give this too much attention.
it's because this is just unverified slop that i doubt was thrown together by a human. Langfuse has ongoing conferences and is still used by many frameworks.
>I had an open source project launching but I found out that I cannot make ShowHN post
FYI, this is because a lot of ShowHN had turned into low quality slop. Not suggesting that it is, but if you do think your project could be slop, consider not posting it here.
This is not just the first stop in promoting your resume project or whatever random open source library. This is for showing _interesting_ projects to the community to get feedback.
I've been working on this project for close to one year and reached the first POC, which is a significant devote of engineering time for me. Though it is still a POC release, I am still proud that I can pull this off:
https://github.com/upwindchange/Autai
I've been working on *Autai* — an open-source desktop app (Electron) that uses AI agents to automate your browser. You just type what you want, and the AI opens a real browser and does it for you.
Beyond that, this software can be used as a normal LLM chat interface, similar to cherry-studio.
*Why this project:*
When I saw the Browser-Use project (the python one), I had a "Wow" moment and tell myself, this would be the future of AI helping people doing everyday task. But when I tried it, it feels very geek to me: you have to use chrome browser via CDP debug mode and the python code runs in a terminal. I was thinking if it is possible to build something that can provide the user an easy to use version of "browser-use" with a nice interface and a nice human in the loop interaction?
I went through a lot of exploration, e.g. build upon cheery studio or lm studio, integrate the python runtime and the whole browser-use project into it. At the very end, every shortcut takes a tradeoff and I decided to write my own from scratch.
So here I am after close to one year of engineering and trials: the first POC release. I wrote the front-end using assistant-ui, back-end using ai sdk. A ton of other packages are used, you can have a look at the package.json file in the project.
Relationship to Browser-Use: You can treat this project as a Typescript rewrite with its own agent, own front-end, etc. More precisely speaking, the CDP tool of browser-use project is rewritten into the Typescript CDP tool in this project. I wrote everything else.
This is THE project in which you won't call an ELECTRON app bloated: because I am using it as a browser, making full usage of its blink + node + CDP debugger engines.
*What it can do:*
- *Browser Automation* — "Add these items to my Target cart", "Book a flight from NYC to SF on Friday", "Fill out this form" — the AI plans the steps and executes them in a real browser session
- *Research Mode* — Ask a question and it searches the web, reads multiple sources, and gives you a synthesized answer. No more 20-tab skimming
- *Multi-session* — Run multiple browser tasks in parallel
- *100+ AI providers, 4,000+ models* — Works with OpenAI, Anthropic, Google, DeepSeek, xAI, Ollama (local), and many more. Bring your own API key
*You stay in control:*
The AI pauses for CAPTCHAs, logins, and payments and hands control back to you. There's a split-view so you can watch everything the AI does in real time.
*Other nice touches:*
- Auto-tagged conversations with search and filtering
- Syntax highlighting, math rendering, Mermaid diagrams in AI responses
- Image and file attachments
- Dark/light mode
- English and Chinese UI
*Project status:*
Autai is in *active alpha development* and evolving fast. I'm heads-down building right now, so issues and feature requests are closed for the alpha phase — they'll open up once it reaches beta. That said, feedback and thoughts are always welcome here in the comments.
MIT licensed. Happy to answer questions about how it works or what's coming next.
>AI agents have changed the scalability properties of basically the entire internet.
Why is GH the only service provider seeing such consistently bad availability then? Everyone has had to scale massively all the time, if GH is choosing moltbots capacity over basic availability for the rest of the humans, they have made the wrong choice.
Some people really abuse the f out of the system in a way optimized to take github down. Like they push every minute or for every commit instead of with certain time intervals (e.g. a single push a few times a day for each repo).
I follow some of the accounts that run 24/7 agent sesssion. Their projects are not even that novel for the number of commits that appear on the profile. Many of the commits have the log of beads, claude session etc (no change to the actual code). Some of them are ports of some projects to another language.
AI surely will increase the productivity, but the waste and noise that some people are willing to commit ....
These aren't normal times though. Low quality submissions have suddenly seriously amplified with the use of LLMs. There has to be a response so project quality and maintainer sanity can be preserved.
Buyers who got an expensive and gaudy pile of shit will never want to admit their pile of shit doesn't smell to themselves.
reply