An AI-powered interview preparation platform designed to help candidates practice, refine, and master technical and behavioral interviews.
- Web Portal: https://interview-ai-lovat-one.vercel.app
- API Health: https://interview-ai-g6fw.onrender.com/api/health
- Hybrid AI Architecture: Utilizes Gemini and HuggingFace for intelligent parsing, alongside robust offline fallback engines and PostgreSQL caching.
- Role-Specific Coaching: Tailored interview questions across multiple disciplines (Software Engineer, Product Manager, Data Scientist, etc).
- Nuanced Feedback Matrix: Answers are evaluated on Relevance, Clarity, Depth, Structure, and Confidence.
- Secure Authentication: Fully fledged JWT email-only authentication flow.
- Dynamic Session Handling: Real-time progress tracking and performance analytics.
- Node.js (v18+ recommended)
- PostgreSQL
-
Clone the repository
git clone https://github.com/prajwal-manjunath20/interview-AI.git cd ai-interview-coach -
Setup the Backend
cd server npm install # Copy the environment example and populate your secrets, database URLs, and API keys cp .env.example .env # Start the backend server npm run dev
-
Setup the Frontend
# From the root directory npm install # Start the frontend Vite server npm run dev
- Frontend: React 19, Vite, Tailwind CSS, Recharts
- Backend: Express, Node.js, PG, Zod
- AI Integrations: Gemini API, HuggingFace Inference API
MIT