Paul Graham essays conversational search

This demo indexes and creates embeddings for 220 essays by Paul Graham in Typesense and uses its conversational search features to enable a natural language chat-based retrieval of the essays. The entire conversation history with the data sources used to generate the response is stored on Typesense.

This implementation uses Next.js 14 App Router for the front end and typesense-js (opens new window) client SDK for sending queries to Typesense.

Live Demo (opens new window) | Source Code (opens new window)

Key Highlights

• Here's (opens new window) how to configure Typesense to create embeddings from documents.
• Here's (opens new window) how to configure an AI model to enable RAG-based responses in conversational search.
• Here's (opens new window) how to call Typesense server using a server action in Next.js.
• Here's (opens new window) how to send conversational search queries to Typesense.
• Here's (opens new window) how to access data-sources used by Typesense to create the responses.