⚡ InterviewPrep.AI

AI-powered mock interview platform that reads your resume, asks questions tailored to you, monitors your focus, and delivers structured feedback — all in one session.

🌐 prepai.thegauravthakur.in  ·  GitHub

---

✨ Features

	Feature	Description
🧠	AI-Personalised Interviews	Gemini reads your resume and crafts questions specific to your stack, experience level, and projects
🎙️	Voice + Code Responses	Answer verbally via speech-to-text or through the built-in code editor for DSA rounds
👁️	Live Proctoring	MediaPipe tracks face and attention — flags anomalies without interrupting your flow
📊	Structured Feedback	Post-session report with strengths, weak areas, and a personalised improvement roadmap
🔁	Adaptive Follow-ups	AI adjusts depth and difficulty in real time based on your answers

---

🏗️ Architecture

┌─────────────────────────────────────────────────┐
│                React Frontend                   │
│   Vite · TailwindCSS · Framer Motion · MediaPipe│
└───────────────────┬─────────────────────────────┘
                    │ REST
┌───────────────────▼─────────────────────────────┐
│              Go Backend (main.go)               │
│   Controllers · Routes · Middleware · Services  │
└──────┬────────────────────────────┬─────────────┘
       │                            │
┌──────▼──────┐            ┌────────▼────────┐
│   MongoDB   │            │   Gemini API    │
│  Sessions   │            │  Q&A · Feedback │
│  Questions  │            │  Resume Parsing │
└─────────────┘            └─────────────────┘

---

🗂️ Project Structure

prepgt/
│
├── client/                          # React frontend
│   ├── public/
│   │   └── mediapipe/               # MediaPipe WASM assets
│   └── src/
│       ├── components/
│       │   ├── Camera/              # MediaPipe face tracking
│       │   ├── Interview/           # Core interview UI
│       │   ├── MicroPhone/          # Speech-to-text
│       │   ├── Speaker/             # TTS playback
│       │   └── Ide/                 # In-browser code editor
│       ├── pages/
│       │   ├── Home/
│       │   ├── Details/             # Resume upload + user info
│       │   ├── Interview/
│       │   └── Report/              # Post-session feedback
│       ├── hooks/
│       └── lib/
│
└── server/                          # Go backend
    ├── main.go
    ├── controllers/                 # Request handlers
    ├── routes/                      # API route definitions
    ├── models/                      # MongoDB schemas
    ├── middleware/                  # CORS, auth, logging
    ├── services/                    # Gemini integration
    ├── db/                          # MongoDB connection
    └── utils/                       # Helpers

---

🧩 Tech Stack

Frontend

• ⚛️ React + Vite — fast dev + optimised builds
• 🎨 TailwindCSS + Framer Motion — UI and animations
• 🎤 Web Speech API — browser-native speech-to-text
• 👁️ MediaPipe — real-time face mesh and attention tracking
• 🖥️ Monaco-style IDE — in-browser code editor

Backend

• 🐹 Go (Golang) — high-performance REST API
• 🍃 MongoDB — session and question persistence
• 🔐 JWT — stateless authentication (coming soon)
• 🤖 Google Gemini API — resume parsing, Q&A generation, feedback

---

🚀 Local Setup

Prerequisites: Node.js v20+, Go 1.21+, MongoDB, Gemini API Key

# 1. Clone
git clone https://github.com/2405Gaurav/prepgt.git
cd prepgt

# 2. Backend
cd server
cp .env.example .env        # fill in your values
go run main.go
# → API running at http://localhost:8080

# 3. Frontend (new terminal)
cd ../client
npm install
npm run dev
# → App running at http://localhost:5173

Backend .env

PORT=8080
GO_ENV=development

MONGODB_URI=""
DB_NAME=""
SESSION_COLLECTION_NAME=""
QUESTION_COLLECTION_NAME=""

GEMINI_API_KEY=""
FRONTEND_URL="http://localhost:5173"

Frontend .env

VITE_SERVER=http://localhost:8080

---

🔐 Auth & Personalisation (In Progress)

The platform is being upgraded from guest-only sessions to a full personalised experience:

• JWT-based authentication — secure login with access + refresh tokens
• User profiles — track your tech stack, experience level, and interview preferences
• Session history — every interview stored, reviewable, and comparable over time
• Growth dashboard — visualise your progress across multiple sessions with score trends
• Personalised difficulty — AI adjusts question depth based on your historical performance
• Interview streaks & milestones — gamified accountability to keep you consistent

---

📈 Roadmap

🔜 Near-term

• JWT auth + user accounts
• Session history and progress tracking
• Growth dashboard with score trends
• Code execution sandbox (Judge0 integration)

🔮 Long-term

• Multi-model support (GPT-4o, Claude)
• Peer mock interviews (P2P matchmaking)
• Company-specific interview packs
• Resume intelligence engine
• Adaptive learning paths between sessions
• Skill gap analytics + curriculum builder
• Institution-level analytics dashboard

---

📜 License

MIT © Gaurav Thakur