Keynote: Ruby Is the Best Language for Building AI Web Apps
summary
Argues RubyLM (the Ruby LLM gem) makes LLM work simple: a Rails chat UI in 1m40s demo shows streaming tokens to the browser, a generator creates chats/messages/models, and streaming jobs append chunks. Claims a one API for 12 providers and supports 800+ models with a model registry and local GPU runtime. Shows a 17‑line concurrent code reviewer using async fibers instead of heavyweight workers, and contrasts clearer Ruby block syntax and Rails integration versus Python/JS tooling.
AI-written summary · about
chapters (20)
- 0:00Ruby for AI web apps
- 1:52Why Ruby for AI
- 2:33LLM libraries compared
- 3:35Ruby LM chat syntax
- 4:39Multi-turn convos in Ruby LM
- 5:22Streaming in Ruby LM
- 6:27Support for structured output and file parsing
- 7:32Gem tools for Ruby LM
- 8:15Provider/model support in Ruby LM
- 9:18Token usage tracking
- 10:43AI features with Rails demo
- 11:48Chat UI in 2 mins with Rails
- 12:55Scaling AI apps with Ruby
- 15:32Multi-agent orchestration
- 17:43Ruby blocks for AI
- 19:08JS vs Ruby LM case study
- 20:12Ruby LM community growth
- 21:57Ruby LM future projects
- 22:41Ruby’s AI moment
- 23:26Q&A
