INTERVIEW.AI is a high-end, production-grade interview practice platform that uses the Gemini 3 Flash model to provide deep, analytical feedback on your interview performance.
- Real-time Voice/Text Practice: Conduct mock interviews using your microphone or keyboard.
- Deep AI Analysis: Get detailed feedback on "substance vs fluff", weaknesses, and a STAR-method improved version of your answers.
- Career Consultant Verdict: A final report with job fit prediction, growth areas, and a personalized improvement plan.
- Session History: Securely save all your practice sessions via Firebase to track your progress over time.
- Modern UI: A polished, responsive interface built with Tailwind CSS and Motion for smooth transitions.
- Frontend: React 18, Vite, Tailwind CSS, Motion (Framer Motion)
- Backend Proxy: Express.js (for serving questions and assets)
- AI Engine: Google Gemini API (
gemini-3-flash-preview) - Database & Auth: Firebase Firestore & Google Authentication
To run this application, you need to configure the following secrets in your environment:
# Required for AI analysis
GEMINI_API_KEY=your_api_key_here
## 📦 Installation & Setup
1. **Clone the repository.**
2. **Install dependencies**:
```bash
npm install- Configure Secrets: Add your
GEMINI_API_KEYto the project secrets. - Run Development Server:
npm run dev
- Personalize: Choose your target Role (e.g., Software Engineer) and Level (e.g., Senior).
- Interview: Answer 5 generated behavioral and technical questions. You can use Voice Mode for a more realistic feel.
- Analyze: Once finished, wait a few seconds for the AI to "bulk evaluate" your entire session.
- Review: Check your individual answer scores and the final "Consultant Verdict" in the History tab.
Built with ❤️ using Google AI Studio.