notebook LLM bot

- Upload pdf
- Find answers you are looking for

TECH STACK

- OpenAI embeddings
- Langchain
- prisma 
- PSQL (for local dev)
- NextJS and AI sdk
- Pinecone DB for similarity search
- Nextauth
- s3 for pdf file storage 

Prerequisites:

• Git: Ensure you have Git installed on your system. You can download it from https://git-scm.com/downloads.

• Node.js and npm (or yarn): These are essential for package management. Install them from https://nodejs.org/en.

• Docker: Docker is required for running the database. Download it from https://www.docker.com/.

• Docker Desktop and Compose Plugin : Gives you one command access to run docker containers

Installation Steps:

1. Clone the Repository:

   git clone https://github.com/shivam-jainn/nbLLM.git

2. Navigate to the Project Directory: Change your working directory to the cloned repository:

   cd NBLLM

3. Install Dependencies: Inside the notebook directory, install the required dependencies using either pnpm, npm, or yarn:

   cd notebook
   pnpm install  # (Recommended for better performance)
   # OR
   npm install
   # OR
   yarn install

4. Start the Database (Docker): Run the following command to launch the database using Docker Compose:

   docker-compose up -d

   The -d flag ensures the database runs in the background.

5. Gather API Keys: You'll need API keys for various services used by the LLM bot. Here's a guide for each:

   • OpenAI API Key:

     • Create an account on OpenAI https://platform.openai.com/signup.
     • Go to your API keys page and create a new API key.
     • Copy and store this key securely.

   • Gemini API Key:

     • If you're using Gemini for any API functionalities, obtain your API key from their platform.

   • Pinecone DB API Key:

     • Create a Pinecone account https://pinecone.org/.
     • Get your API key from your project settings and store it securely.

   • AWS Environment Variables:

     • If the bot interacts with AWS services (e.g., S3 for file storage), create and configure the necessary environment variables in your system or a .env file. You can find instructions on setting environment variables specific to your operating system.

   • Google OAuth API Key:

     • If Google OAuth is used for user authentication, create a project and API credentials in the Google Cloud Console https://developers.google.com/workspace/guides/create-project. Follow the on-screen instructions to obtain your API key.

   • GitHub OAuth API Key:

     • Similar to Google OAuth, create a GitHub application and obtain an API key if the bot uses GitHub for authentication.

   • NextAuth Secret:

     • Create a secret to use with NextAuth. You can use the following command in your terminal:

     openssl rand -base64 32

     This will generate a random 32-character string suitable for a secret.

Configuration:

• Create a .env file (optional): You can create a .env file in the root of your project to store environment variables securely (e.g., API keys). This file should not be committed to version control.

Verification:

Once you've completed the installation and configuration steps, you can typically run the bot using a command like npm run dev or yarn dev or pnpm run dev. This will start the development server and allow you to access the bot interface in your browser (usually at http://localhost:3000).

Additional Notes:

• Right now , this bot only supports one pdf at a time