StudyAssistant with Google Gemini

This project is a web-based AI assistant that allows you to upload a PDF document and ask questions about its content. It uses a Retrieval-Augmented Generation (RAG) pipeline powered by Google's Gemini Pro model and LangChain to provide intelligent and context-aware answers.

---

Features

• Interactive UI: A clean and user-friendly interface for uploading files and chatting.
• PDF Upload: Supports uploading PDF documents via button click or drag-and-drop.
• Document Processing: Extracts text from the PDF, splits it into manageable chunks, and creates a searchable vector index.
• Conversational Q&A: Ask questions in natural language and receive answers generated by a powerful AI model based on the document's content.
• Powered by Gemini: Utilizes Google's gemini-2.5-pro for high-quality responses and embedding-001 for text embeddings.

---

Tech Stack

• Backend: Python, FastAPI
• AI/ML: LangChain, Google Generative AI (Gemini), FAISS (for vector storage)
• Frontend: HTML, CSS, JavaScript
• PDF Processing: PyPDF2

---

How It Works

The application follows a Retrieval-Augmented Generation (RAG) architecture:

1. PDF Processing: When a user uploads a PDF, the backend extracts all text content.

2. Text Chunking: The extracted text is split into smaller, overlapping chunks to ensure semantic context is preserved.

3. Vector Embedding: Each text chunk is converted into a numerical vector representation (embedding) using Google's models/embedding-001.

4. Vector Store: These embeddings are stored in a local FAISS vector store, which allows for efficient similarity searches.

5. Question Answering:

   • The question is converted into an embedding.
   • The FAISS store is searched to find the text chunks with embeddings most similar to the question's embedding.
   • These relevant chunks (the context) and the original question are passed to the Gemini Pro model via a carefully crafted prompt.
   • The model generates a final, human-like answer based on the provided context.

---

Setup and Installation

Prerequisites

• Python 3.9+
• A Google API Key with the "Generative Language API" enabled (available from Google AI Studio).

2. Create a Virtual Environment

# Windows
python -m venv venv
venv\Scripts\activate

# macOS/Linux
python3 -m venv venv
source venv/bin/activate

3. Install Dependencies

Create a requirements.txt file with the following:

fastapi
uvicorn[standard]
python-dotenv
google-generativeai
langchain-google-genai
langchain
pypdf2
faiss-cpu

Install everything:

pip install -r requirements.txt

4. Add Your API Key

Create a .env file in the main project folder and add:

GOOGLE_API_KEY="PASTE_YOUR_SECRET_API_KEY_HERE"

5. Run the App

uvicorn app:app --reload

Then open your browser and go to: http://127.0.0.1:8000 live demo : https://studyassistant-dzq4.onrender.com/

---

Project Structure

.
├── app.py              # The heart of our backend server
├── templates/
│   └── index.html      # The main (and only) HTML page
├── static/
│   ├── style.css       # Makes everything look pretty
│   └── script.js       # Powers all the interactive features
├── .env                # Where your secret API key lives
├── requirements.txt    # A list of the Python tools we need
└── README.md           # You're reading it!

---