RubyLLM: Making AI Development Beautiful Again
summary
RubyLLM is a Ruby-native LLM library that provides one API across 11 providers with 620 models in the model registry, cost and capability tracking, and progressive-disclosure helpers for chat, embeds, moderation, tools, and structured output. It demos a Rails generator that scaffolds a chat UI and database models and streams tokens to the browser, scales by swapping Puma→Falcon and using fiber-based sync jobs, and boasts 1,700+ GitHub stars in its first week.
AI-written summary · about
chapters (22)
- 0:00Introduction
- 1:13AI development and Ruby
- 2:00AI complexity myths
- 3:27Ruby LM prompt templates
- 4:32Unified API for models
- 5:58LangChain’s complexity critique
- 7:01Ruby LM’s guiding principles
- 8:05Simple and complex AI use cases
- 9:10Models as commodities
- 10:16Model and provider support
- 11:21Switching models mid-conversation
- 12:04Convention over configuration
- 12:24Progressive disclosure features
- 13:06Custom parameters and function calling
- 13:48Structured output with schema
- 14:31Rails integration and chat UI
- 16:22Building chat UI
- 17:50Rails integration architecture
- 18:33Scaling AI apps
- 21:03Performance optimization
- 21:27Ruby LM adoption and impact
- 23:13Call to action for Ruby
related talks
21mRubyLLM
28mKeynote: Ruby Is the Best Language for Building AI Web Apps
28mGetting More Out of LLMs for Ruby Developers Justin Wienckowski, Jake Gribschaw We're all hearing about LLM tools that are changing on a daily, if not hourly, basis
01h 04mBuilding LLM powered applications in Ruby
33mLLM Telemetry as a First-Class Rails Concern
43mBuilding AI agents in Ruby
16mIs that LLM feature any good?
27mGoing Postel