Your AI-powered study companion for CBSE Class 9 and 10 students in India.
π Visit the live site: cbseai.live
CBSE AI is designed to help students of CBSE Class 9 and 10 get instant, reliable answers to their academic questions. Leveraging Gemini, it provides support across all subjects, anytime and anywhere[^1].
- β‘ Instant Answers: Get immediate, accurate responses to your questions across all CBSE subjects[^1].
- π 24/7 Availability: Access your study companion whenever you need help β day or night[^1].
- π Optimization at its peak: Built with Next.js and LangChain.js for a fast, interactive, and scalable experience.
| Layer | Technology Used |
|---|---|
| π Frontend | Next.js |
| π§ AI Logic | LangChain.js |
| π Auth | Auth.js with Google OAuth |
| π¦ Deployment | Vercel |
| π§ Vector DB | Pinecone |
| π€ AI Model | Gemini |
+---------+ +----------------+ +-------------+
| Browser | <---> | Next.js Server | <---> | LangChain.js|
+----+----+ +--------+-------+ +------+------+
| | |
| Auth.js + | |
| Google OAuth2 | |
| | |
| +-------v-------+ +------v------+
| | Gemini | <---> | Pinecone |
| +---------------+ +-------------+
π¬ Frontend: Users interact with a chat interface built with React to submit questions.
π§ Backend: Questions are processed via LangChain.js, which orchestrates calls to the Gemini API and returns structured, relevant answers by utilizing a vector database called Pinecone.
app/api/ β Backend Stuff
app/api/chat/route.ts β API route for handling chat and AI logic
app/chat/page.tsx β Main chat interface (frontend)
components/ β UI components from shadcn/ui
- π¦ Node.js (v20+ recommended)
- π§βπ» A code editor (Zed recommended)
- π₯ Clone the repository:
git clone https://github.com/arjav0703/cbse-ai.git
cd cbse-ai- π¦ Install dependencies:
npm install-
π Set up environment variables: Create a
.env.localfile in the root directory and add your environment variables. You can use the.env.examplefile as a reference. -
π Start the development server:
npm run devπ Start the production server:
npm run build && npm run start- π Open your browser:
http://localhost:3000