Mine is https://github.com/schappim/macOCR

For me it is an old version of FlyingLogic before they went subscription only.

This is awesome

For me it was the open source e-ink display TRMNL[2] which I learned about via ATP and Snazzy labs[1].

1. https://youtu.be/eIcZZX10pa4?si=deMGwccwwO4wJdBe

2. https://usetrmnl.com

I’ll go first:

  - PiHole (for network wide adblocking)
  - Home Assistant for home automation.

I had a similar experience when I bought a pressure washer[1] that required an app. Some things just shouldn’t need one.

1. https://www.kaercher.com/au/home-garden/pressure-washers/sma...

They want you to fumble around with a phone app to adjust the machine, while you're presumably going to be at least somewhat wet from its use? WTF.

Drawing from my experience, modern technologies like Hotwire and Turbo Frames—when integrated with background jobs and Action Mailbox for handling inbound emails—alongside mature tooling, create a surprisingly effective (and enjoyable) platform for developing AI applications.

Isn't that just tooling around integrating with any API?

I'm not sure what you mean.

I attempted to implement this on the front end of my e-commerce site, which has approximately 20,000 products (see gist [1]). My goal was to enhance search speed by performing as many local operations as possible.

Biggest impact in performance was by moving to dot products.

Regrettably, the sheer size of the index of embeddings rendered it impractical to achieve the desired results.

1. https://gist.github.com/schappim/d4a6f0b29c1ef4279543f6b6740...

This looks nice. I also played on the weekend with Vue and Transformer.js to build the embeddings locally. See https://github.com/alexanderop/vue-vector-search

Your username suggests that you might already have an answer to this.

Here is a basic implementation of this in Ruby: https://gist.github.com/schappim/ad8b4953486617c7f813751c8ee...