Pinecone and Google Generative AI Integration

This project demonstrates how to integrate Pinecone, a vector database, with Google Generative AI to perform operations such as embedding generation, data insertion, querying, and retrieving index statistics.

Project Structure

Files in the Folder:

1. Insert-data.js

   • Inserts embedding vectors into a Pinecone index.
   • Generates embeddings for a given content using the Google Generative AI text-embedding model.

2. search.js

   • Queries the Pinecone index using a vector generated from Google Generative AI embeddings.
   • Retrieves the top K matching vectors with their metadata.

3. Find.js

   • Retrieves and prints information about the Pinecone index, including namespace details and statistics.

---

Prerequisites

1. Node.js

   • Ensure you have Node.js installed on your system.

2. Install Dependencies

   • Install the required libraries:

     npm install @pinecone-database/pinecone @google/generative-ai

3. API Keys

   • Obtain and set the following API keys:
     • Pinecone API Key: Available in your Pinecone account.
     • Google Generative AI Key: Available in your Google Cloud Console.

---

Setup

1. Environment Variables

   • Store API keys securely by creating an .env file in the project root:

     PINECONE_API_KEY=your-pinecone-api-key
     GOOGLE_AI_API_KEY=your-google-ai-api-key

2. Folder Structure:

   /project-folder
   ├── Insert-data.js
   ├── search.js
   ├── Find.js
   ├── README.md
   ├── .env