Building LLM powered applications in Ruby
summary
A Ruby app uses LangChain RB and the OpenAI chat/completions API to run a simulated e‑commerce backend called nerds and threads, wiring tools for customer, inventory, payment, shipping and email into function calls. The demo shows RAG-style retrieval, embedding searches into a SQLite vector workflow, and agents executing order, refund and inventory flows end‑to‑end; most runs succeed (author reports about nine times out of ten), with failures attributed to agent generality and hallucinations.
AI-written summary · about
chapters (9)
- 0:00Introduction and background
- 3:20What are LLMs and generative AI
- 8:31AI agents and business logic
- 15:34Retrieval Augmented Generation explained
- 27:36Live coding demo setup
- 30:24Building the e-commerce AI assistant
- 49:58Testing order processing and returns
- 57:23Why Ruby for LLM applications
- 1:00:16Project insights and closing remarks
related talks
43mBuilding AI agents in Ruby
01h 59mLLM applications with LangChain
24mRubyLLM: Making AI Development Beautiful Again
21mRubyLLM
28mKeynote: Ruby Is the Best Language for Building AI Web Apps
30mBuild an AI agent in 15 min
27mI know Kung Fu! RubyGems for AI
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