Vector databases and large language models (or LLMs) enable fast prototyping of systems that were incredibly difficult to build in the past. In this video, you'll learn how to leverage the power of AI through Semantic Search to query a dataset of books producing recommendations based on user inputs. This project is packaged in a NextJS application and a Python data pipeline. You can build this locally or on Replit using our Repl!
Live demo: bookrecs.weaviate.io/
GitHub: github.com/weaviate/BookRecs
Repl: replit.com/@Weaviate/AI-Book-...
Weaviate Forum: forum.weaviate.io/
Weaviate Community Slack: weaviate.io/slack
Kaggle Dataset: www.kaggle.com/datasets/dylan...
Music: Mixkit - Driving Ambition
Introduction 0:00
Accounts & Environments 1:43
Loading Data 4:55
Semantic Search Query 7:15
NextJS App 8:39
Input Form 9:05
Defining the API 10:32
Recommendation Grid 11:15
Modal View 12:10
Outro 13:05