Recent submissions

Comment · 25 days ago
Imagine a worker in a car factory who inspect the car's body. Pictures from all angles and AI inspection is faster, cheaper more thorough and works 24 hours a day. And that is only 1 scenario.
Comment · 1 month ago
Reasons or details?
Comment · 1 month ago
Workerman does something else.
Comment · 2 months ago
Why is it so hard for many people to accept, that this is a solution for a specific problem of shopify? They did not say that Redis is bad and MySql is good. They only a solve their problem.
Comment · 2 months ago
Cool idea, but please make it better!
Comment · 2 months ago
Comment · 3 months ago
"Nothing stops you from using sky-400 and blue-400 in the same project. Consistency still depends on your discipline, not on the framework."

And how did plain css solve that? You're mixing up a few issues there.

Comment · 3 months ago
Or:

  .button {
    padding: var(--gap-s) var(--gap-m);
    background-color: var(--color-indigo);
    color: var(--color-white);
    font-weight: 700;
    border-radius: var(--gap-s);
    &:hover, &:focus {
        background-color: var(--color-indigo-hover);
    }
  }
Fair comparison (i really don't like this too):

  .button { padding: var(--gap-s) var(--gap-m); background-color: var(--color-indigo); color: var(--color-white); font-weight: 700; border-radius: var(--gap-s);    &:hover, &:focus { background-color: var(--color-indigo-hover); } }
;-)
Comment · 3 months ago
if you don't like this.

  .button {
      @apply py-2 px-4 bg-indigo-500 text-white font-semibold rounded-lg hover:bg-indigo-700 focus:bg-indigo-700;
  }

why not write it like this?

  .button {
    @apply
      py-2
      px-4
      bg-indigo-500
      text-white
      font-semibold
      rounded-lg
      hover:bg-indigo-700
      focus:bg-indigo-700
    ;
  }
Comment · 3 months ago
I've had the same experience. Tried it and use in 2 hours more token than in a whole week. Normally i got 80 - 90 % cache hit rate.
Comment · 3 months ago
And how much token and time you use to solve a problem? The price alone doesn't mean anything.

Example DeepSeek-V4-Pro (high) needs 10 times more token then GPT 5.5 (medium) and can compete only with the price.

The real price saver are the cache prices, the ting, that nearly nobody has on their radar.

Comment · 3 months ago
Ockbench: https://ockbench.github.io/ show token usage and accuracy, but you have to compare it with other stats (prices and speed).
Comment · 3 months ago
"Open won"... to be fair cause "google paid it".
Comment · 3 months ago
Comment · 3 months ago
Costs?

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton