An AI-powered chat application with image analysis capabilities.
As shown above, BlazerAI’s guardrails are working effectively with Gemini. When an answer is correct, the guardrail policy acknowledges it and provides guided prompts that encourage deeper exploration of the topic. This approach helps students go beyond surface-level responses, allowing them to learn multiple related concepts and gain a more thorough understanding.
BlazerAI/
├── client/ # Frontend (React + Vite)
│ ├── src/
│ ├── public/
│ ├── package.json
│ └── .env
└── backend/ # Backend (Node.js + Express)
├── index.js
├── package.json
└── .env
- Navigate to client folder:
cd client- Install dependencies (already installed):
npm install-
Configure
.envfile with your API keys (already configured) -
Start development server:
npm run devThe frontend will run on http://localhost:5174
- Navigate to backend folder:
cd backend- Install dependencies:
npm install-
Configure
.envfile with your credentials -
Start development server:
npm startThe backend will run on http://localhost:3000
- AI-powered chat using Google Gemini
- Image upload and analysis
- User authentication with Clerk
- Real-time chat streaming
- Markdown support for responses
- Upload button on dashboard
- Guard rail Policy
- React 19
- Vite
- React Router
- TanStack Query
- ImageKit for image handling
- Google Generative AI
- Node.js
- Express
- MongoDB
- ImageKit
- TanStack
- Clerk