🧠 AI Study Assistant — Chat with Your PDFs

Upload • Understand • Learn

An intelligent AI web application that lets you upload PDFs and chat with their content — powered by Cohere embeddings, Gemini AI, PDF.co, and Supabase vector search.

---

🚀 Overview

AI Study Assistant (also known as Chat with PDFs) is a full-stack AI project that allows users to upload a PDF, process its content, and ask natural-language questions about it.

When a user uploads a PDF:

1. The text is extracted using the PDF.co API.
2. The content is split into smaller, manageable chunks.
3. Each chunk is converted into an embedding using Cohere’s embedding model.
4. The embeddings and text are stored in a Supabase vector database.
5. When the user asks a question, the app retrieves relevant chunks and sends them to Gemini AI to generate an accurate, context-aware answer.

This follows a Retrieval-Augmented Generation (RAG) pipeline — meaning your chatbot answers directly from your uploaded PDFs, not from random AI knowledge.

---

🧩 Features

✅ Upload and process any PDF
✅ Extract text using PDF.co API
✅ Generate embeddings via Cohere API
✅ Store embeddings efficiently in Supabase
✅ Chat with your uploaded document using Gemini AI
✅ Contextual, accurate, and real-time answers

---

🏗️ Architecture

                ┌──────────────────────────┐
                │        Frontend (UI)     │
                │  Next.js + Tailwind CSS  │
                │  ─────────────────────── │
                │  • PDF Upload Form       │
                │  • Chat Interface        │
                └─────────────┬────────────┘
                              │
                              ▼
                ┌──────────────────────────┐
                │        API Routes        │
                │  (Next.js App Router)    │
                │  • /api/upload → Extract, Embed, Store │
                │  • /api/chat → Retrieve, Answer        │
                └─────────────┬────────────┘
                              │
                              ▼
                ┌──────────────────────────┐
                │       Supabase DB        │
                │ (Vector Storage & Search)│
                │   • Stores embeddings    │
                │   • Performs similarity  │
                └─────────────┬────────────┘
                              │
                              ▼
                ┌──────────────────────────┐
                │        AI Models         │
                │ ──────────────────────── │
                │ • Cohere Embeddings (v3) │
                │ • Gemini 1.5 Flash (LLM) │
                └──────────────────────────┘