Node.js PDF Knowledge Base

Tech Stack

• JavaScript
• Node.js
• Express.js
• MongoDB
• pdfReader
• Embedding API (OpenAI)
• Thunder Client (API CLIENT)

Local Development

Start developing locally.

Step-1

clone this repo

git clone https://github.com/Swoyam1/KnowledgeBase-RAG.git

Step-2

Install all dependencies

# install server side deps
npm install

Step-3:

Create a SEARCH INDEX in MONGODB ATLAS

Step-4:

Create a .env file in root folder and write these code

MONGO_URL = "PROVIDE YOUR MONGODB ATLAS URL"
OPENAI_API_KEY = "PROVIDE YOUR OPENAI API KEY"

Step-5: Starting the server

Finally to start the server execute this script

npm run dev

After starting the server it should be running on http://localhost:7000

API

/document

• POST : Post query to add vector embedding of PDF to the database

/query

• POST : Post query and get the answer to the query in response

# post query body element
{
  "query" : "PROVIDE YOUR QUERY"
}