A fully functional Google NotebookLM-inspired web application built using Next.js, TypeScript, and Google Gemini API. This project allows you to upload documents, generate embeddings, and interact with your data through AI-powered chat, all with in-memory search (no database required).
- 📂 Upload PDF & documents and process them into searchable chunks.
- 🧠 AI-Powered Q&A using Google Gemini embeddings.
- 🔍 In-memory vector search (simple and fast, no external DB needed).
- 🖥️ Modern UI built with React, TailwindCSS, and Next.js 14 App Router.
- ⚡ Streaming responses for a ChatGPT-like experience.
- Frontend: Next.js (React + TypeScript), TailwindCSS
- Backend: Next.js API Routes, Node.js
- AI & Embeddings: Google Gemini API
- Vector Search: Lightweight in-memory search engine
-
Clone the repository:
git clone https://github.com/anirudh7065/Google-NotebookLM-Clone.git cd Google-NotebookLM-Clone -
Install dependencies:
npm install # or yarn install -
Set up environment variables: Create a
.env.localfile in the root directory and add the following:GEMINI_API_KEY=your_google_gemini_api_key_here
- Get your Gemini API Key from Google AI Studio.
-
Run the development server:
npm run dev # or yarn devVisit
http://localhost:3000to see the app in action.
├── app/ # Next.js App Router (routes, API endpoints)
├── components/ # Reusable React components
├── lib/ # Utilities (AI calls, embeddings, vector search, etc.)
├── public/ # Static assets (icons, images, screenshots)
├── types/ # TypeScript type definitions
├── .env.local # Environment variables (not committed)
├── package.json # Dependencies and scripts
└── README.md
- Upload one or more PDFs.
- The app will split documents into smaller chunks and generate Gemini embeddings.
- Ask questions in the chat – answers will be retrieved from your documents using in-memory vector search.
Contributions are welcome! Feel free to fork the repo, create a branch, and submit a PR.
This project is licensed under the MIT License.