Document Analyzer

An AI-powered application that analyzes PDFs using the Retrieval-Augmented Generation (RAG) technique to deliver accurate, context-aware answers from your documents.

It simply works by uploading a PDF and prompting the LLM questions about the document.

Built with:

• Laravel 12
• VueJS
• InertiaJS
• PrismPHP

How It works (Client Side)

• User uploads a document (PDF)
• The system processes the document
• Once processing is complete, the user can prompt the LLM for queries related to the document.
• Users can start new conversations or continue from previous sessions.

How It Works (Server Side)

• User uploads a PDF
• Text is extracted from the PDF using Smalot PDF Parser package.
• The extracted text is segmented into smaller chunks for efficient processing and to meet the embedding model context.
• Each chunk is sent to an embedding model to generate vector representations.
• The resulting vectors are stored in a vector database for retrieval.
• When the user submits a query, the system performs a similarity search in the vector database using the RAG (Retrieval-Augmented Generation) technique to find relevant context.
• The retrieved context is passed to the LLM, which generates a context-aware response.
• The LLM’s response is returned and displayed to the user through a chatbot interface.

Notable Features

• Manager Design Pattern - The system uses a Manager pattern, allowing easy configuration and switching between different Vector Databases and LLMs.

  • Vector Database: Qdrant
  • LLM: Gemini

• Queues - Each text chunk is enqueued as an individual job. Batchable jobs are grouped together and processed in parallel to optimize throughput and performance.

• Job Middleware - To comply with the LLM’s rate limits, a middleware enforces a rate limit of 100 requests per minute (RPM) for job execution.

• Reverb - A real-time communication layer that enables live tracking of job progress and interactive chatbot updates. It uses event-driven architecture with WebSockets to broadcast job state changes, logs, and chatbot messages as they occur.

Installation

• Copy this repository

• Copy the .env.example to .env

• Setup the database connection

• Install Composer

  composer install

• Generate the application key

  php artisan key:generate

• Migrate and seed the database

  php artisan migrate:fresh --seed

• Run the project

  php artisan serve