Abstract
Consumed by the AI wave this year, I pondered what the future holds for Ruby in the world of ever-growing AI/ML needs. Having to have just completed a massive client project building a keyword search with Elasticsearch, I dove into the current landscape of semantic search capabilities and the vector search databases. This led me to building Langchain.rb, an “original Langchain”-inspired open source library for building LLM-powered applications. It supports use-cases such as semantic search, chat bots, Q&A, prompt management, and even experimental agents. I’m going to demo the library and talk about the lessons I’m learning along the way.
As a Ruby community we should be able to adapt to constantly evolving needs and build our own sets of tools and methodologies suitable for our stack.
Bio
Andrei has been a software engineering professional for almost 13 years. Among many others he’s been fortunate to make his impact at Spree Commerce (Acquired by First Data), WeddingWire (merged with The Knot), FiscalNote (IPO), National Public Radio, and USA Today. He currently runs a software dev firm, and serves as an Architect/Engineering Manager/Fractional CTO on the client projects. In his free time he enjoys playing tennis and going on long runs while listening to podcasts.
28 июл 2024
