AI and LLM-powered apps are all the rage in today’s software development world, and Node.js is no exception.
The average employee spends 3.6 hours daily searching for organizational information (source: Coveo). Now, with the rise of generative AI, it has never been easier to build a question-answering system based on your organization's own data, that can provide accurate answers even to questions phrased freely and vaguely - all in Node.js!
In this talk, we'll go over the elements of RAG (Retrieval Augmented Generation) and show how to build a RAG-based bot in Node.js using Langchain or LlamaIndex, a vector database (e.g., Pinecone, Chroma, Redis), and the OpenAI API. We will then demonstrate how to expose it as a Slack app for all employees to use, or as a REST API to build more software on!
Roy Mor is a software engineer and architect by day and a musician by night (or vice versa). He has worked at early-stage startups, multinational corps, mid-size companies and as an independent consultant, and has been using Node.js since 2015. He has worked on data engineering, cloud services, APIs, test automation, front end UI and embedded firmware, writing some beautiful bugs. His current interests and activities are building LLM-powered apps and using AI agents to drive software forward.