Saguaro AI

This project was inspired by the fact that there’s a ton of baseball news, and I just cannot keep up. My goal was to create something that could read news stories I might care about and then summarize them in a way that both retains the content integrity and exposes the content to me in a way that’s still beneficial for the publisher.

I’m not all the way there yet — this project is very much a WORK IN PROGRESS.

That being said, it’s already been useful to me! I’ve got a laundry list of things that need doing still, but the fact that it’s already been able to deliver value to my real life is pretty neat. Here’s an example of what a story overview looks like currently:

Here’s a screenshot of the homepage:

Technology Used

• Ruby on Rails 8, which has been an absolute pleasure. I’ve been using Rails professionally forever, but I’ve really enjoyed the fun new stuff Rails has released.
• Hotwire/Turbo for a fast and responsive frontend.
• StimulusJS for easy frontend interactivity.
• TailwindCSS and DaisyUI for easy and consistent styling.
• Kamal for deployment to a DigitalOcean droplet and managed PostgreSQL database. This step in particular was a breeze compared to a normal Rails deployment.
• OpenAI for the LLM.