C.L.A.R.A – Conversational Learning AI for Recruitment Assistance

C.L.A.R.A (Conversational Learning AI for Recruitment Assistance) is an AI-powered mock interview assistant that helps students and job seekers practice job interviews in a realistic environment.
It uses voice interaction, AI feedback, and real-time streaming to simulate interview scenarios and provide constructive feedback.

👉 Live Demo: clara-ai-interviewer.vercel.app

---

✨ Features

• 🎙️ Voice-based Interview Practice – Speak naturally, get real-time AI responses.
• 🤖 AI Feedback – Instant analysis on your answers (tone, clarity, relevance).
• 🔄 WebRTC/WebSocket Streaming – Low-latency voice streaming with reconnect support.
• 🧩 Next.js + Node.js Backend – Fast, scalable, and production-ready.
• 🎨 Modern UI – Built with TailwindCSS and Framer Motion.
• 🔐 Authentication – Secure sign-in/sign-up flow.
• 📊 Feedback Dashboard – Review transcripts and performance insights.

---

🛠️ Tech Stack

• Frontend: Next.js 14 + React
• Styling: TailwindCSS + custom theme tokens
• Fonts: Google Fonts (Mona Sans)
• Auth: NextAuth.js (Email/Password)
• Backend: Node.js + Express
• Database: MySQL (eventually with Prisma ORM)
• AI/Voice:
  • Whisper (Speech-to-Text)
  • TTS (Text-to-Speech)
  • VAD (Voice Activity Detection)
  • LangChain / LiveKit / Pipecat (for orchestration & real-time streaming)

---

📂 Project Structure

.
├── app/                 # Next.js App Router pages & layouts
│   ├── layout.tsx       # Root layout with global styles & dark mode
│   ├── page.tsx         # Landing / auth entry
│   └── ...
├── components/          # UI components (buttons, cards, etc.)
├── styles/
│   └── globals.css      # Tailwind + theme tokens + dark mode
├── public/              # Static assets (patterns, logos, etc.)
└── README.md