I am using RubyLLM for quite some time and I am in love with the API design. If someone wants to see how this looks in a real project including custom tools, you can have a look at the SerpTrail project on GitHub.
The Chat model still is just:
class Chat < ApplicationRecord
acts_as_chat model: :llm_model
end
YouTube is replacing my Netflix now, honestly. But I am not happy it being just an algo game, so I am building tubeandchill.com to find good creators, get video tips by newsletter, and more... (tell me what you want to see there, please).
Yes, the distribution is super hard. I recently changed how I blog and building entire platform around that called LakyAI (https://lakyai.com). The idea is being able to run multiple blogs in parallel with crosslinking, sharing resources, reposting to platforms that give you distribution, repurposing the blogs entirely, and more. I am at the beginning but if someone is interested in some novel ideas just drop me an email at [email protected].
I am working on LakyAI, https://lakyai.com, marketing platform to build brands. It's a mix of WordPress and Buffer in a way. Great for people that want to blog but also get some distribution and social media help. If you want to try it, please DM me at x.com/strzibnyj
I know well what you are talking about since I did something similar, but I finally moved to Docker with Kamal (except one project I still have to move). The advantage of Docker's reproducibility is to have a peace of mind when comes to rollbacks and running exact versions due to system dependencies. If anyone is curious I wrote Kamal Handbook to help people adopt Kamal which I think brings all the niceness to Docker deployment so it's not annoying.
Very nice and minimal. I am going the opposite direction. I am cancelling my static generated blogs and moving everything to LakyAI which will help me to manage everything from a single place. Not quite ready for prime time tho.
Ruby is plenty fast these days (people usually think Ruby didn't get faster but it did). And if we figure out how to switch to async/fibers with let's say Falcon, we are in a good place :)