ohadron

This is extremely cool to the point where I feel dangerous even knowing that the option to do this exists.

Favorite quote: Anything that makes engineering harder makes agentic engineering much harder.

Obviously things will get expensive quick, but the main thing for me would be not dealing with the context switch every time I leave the agent to do stuff on it's own.

Feedback loops for prototyping could become even quicker.

This is great. Agentic coding at 600+ tokens/sec is going to be a radically different beast. Coming soon-ish?

The maximum theoretical size for a zip archive is 16 exabytes (2^64 bytes). It's free if you have where to store it.

But why

> Each step moves further from "how do we build better models?" toward "how do we monetize the models we have?"

I don't think OpenAI launching ChatGPT Apps and Atlas signals they're pivoting.

It's just that when you raise that much money you must deploy it in any possible direction.

LLMs would be amazing for this

This is a terrific idea and could also have a lot of value with regards to accessibility.

The web developer / web page ratio in 2025 is for sure way lower than it was in 1998.

Took a while but Wix / Webflow / SquareSpace / Wordpress did end up automating a bunch of work.

The tag line is... scary. I don't want anyone or anything controlling my computer.

Maybe it can have it's own computer? like an employee?

For one thing, it's way faster than the OpenAI equivalent in a way that might unlock additional use cases.

That’s actually a good prompt

Took a bit to run but now my iPhone feels much faster. Thanks!

PSA: Don't shorten URLs if you don't have to because if the shortener ceases working your link is dead.

I was skeptical but the demo is incredible (https://beta.retellai.com/home-agent)

It's quite evident after you read a few. Typically convoluted and fancy English.

In other words: Upon perusal of a mere handful, the prevailing characteristic becomes abundantly clear. The prose predominantly revels in a labyrinthine complexity, intertwined with a flourish of ornate lexicon.

Cool! Feature requests:

1. physical synthesis of typewriter sounds rather than using sound files.

2. Simulating key jams, so if you type too quickly further key presses are inhibited.

Same. Can't imagine why they've been using it for so long instead of building a reasonable UX for this use case. I think this will cost them a bunch of traction.

I couldn't find any evidence for this font having particularly good legibility (such as a Rapid Serial Visual Presentation (RSVP) test), but my hunch is that it's not that great.

It was probably optimized for low-res screens and specifically for monospace usage.

Kerning looks quite off in the Google fonts specimen (https://fonts.google.com/specimen/B612?query=b612)

Please don't build up my hopes, I'm not sure I'll be able to withstand additional disappointment.

This is how Waze started.

What's the value proposition here? It's just the free ChatGPT app only you're paying for all of us parasites who don't want to create an account?

Interesting. A few followup thoughts:

- Can you share this analysis across a longer time span?

- Maybe there's a way to normalize it to general activity on HN? It probably hasn't gone down, but it would be cleaner to do that.

Good question. I think Goodreads put a bit more emphasis on audience reviews and sentiment other than just trying to outright sell it.

But yes, the difference is not huge.

I've just started reading Project Hail Mary, and this is more or less how it starts.

https://www.goodreads.com/book/show/54493401-project-hail-ma...