The State of the Ruby AI Toolbox
summary
Catalogues three production-ready Ruby gems and explains what each actually does and how they plug into a Rails app. Demonstrates a Rails 8 demo app (SQLite, OpenAI key) that Ray (framework agnostic) wraps chat completions and tools but leaves persistence to you, Ruby LLM (Rails-first, DB-backed messages) auto-generates models/controllers and is already used in production at Core Pilot, and Rafe (Rails engine with admin UI) ships views, prompts, an audit UI and evaluation tooling — demo code runs locally “works on my machine.”
AI-written summary · about
chapters (10)
- 0:00Introduction
- 4:23Why build AI in Ruby
- 6:33LLM integration basics
- 11:29Demo app and Ruby AI tools
- 12:10Ray framework-agnostic library
- 16:52Ruby LLM Rails gem
- 21:27Rafe Rails engine
- 25:41Other libraries and future
- 27:49Practical AI features in Ruby
- 28:31Q&A
related talks
21mRubyLLM
28mKeynote: Ruby Is the Best Language for Building AI Web Apps
27mI know Kung Fu! RubyGems for AI
08mRuby startup demo – Zach Kamran
37mPatterns of Application Development Using AI
43mBuilding AI agents in Ruby
19mBuilding a Rails Engine From Scratch: Active Storage Dashboard
30mThoughtful AI for the Rubyist