🧠 FaceCode — Adaptive AI Coding Platform

FaceCode is a next-generation intelligent coding environment that dynamically adapts its difficulty and provides real-time AI interventions based on your facial expressions, keystrokes, and live physiological confidence metrics.

By unifying a robust code execution sandbox with cutting-edge computer vision and natural language processing, FaceCode acts as a personalized, empathetic AI Mentor that knows precisely when you're frustrated, focused, or coasting.

---

✨ Core Features

1. Real-Time webcam Telemetry

   • Emotion Recognition: Utilizes DeepFace and FER to read live emotional states (e.g., Happy, Angry, Fear, Neutral).
   • Confidence Estimation: Computes a live "Focus & Confidence" percentage score utilizing MediaPipe head-pose approximations and facial landmark stability.
   • Keystroke Dynamics: Analyzes WPM (Words Per Minute), CPM (Characters Per Minute), and backspace frequency to track hesitation or coding flow blocks.

2. Adaptive Difficulty Engine

   • Calculates a true "Skill Rating" based on solve times, compilation errors, and emotional distress levels during a problem.
   • Automatically throttles or increases the difficulty of subsequent LeetCode challenges fetched from the backend to match the user's flow state optimally.

3. Context-Aware AI Mentor Interventions

   • If severe frustration (high anger/sadness + high cognitive load + compilation errors) is detected for an extended period, the AI Mentor (powered by LLMs like Groq/Llama3) triggers a supportive intervention modal.
   • Generates contextual hints without giving away the direct answer.

4. Advanced Analytics Dashboard

   • Glassmorphism UI: A gorgeous, strict monochrome + neon pastel interactive UI built with React.
   • Efficiency Trends: Live <LineChart> tracking Completion Time vs. Accuracy.
   • Topic Mastery: Dynamic CSS Grid Heatmap clustering coding topics by Weak/Moderate/Strong success rates.
   • Emotional Insights: Live <PieChart> distributing your session's mental state footprints.

---

🛠 Tech Stack

Frontend

• Core: React 19, Vite, Javascript
• UI/Styles: Vanilla CSS (Glassmorphism design system), Lucide-React icons
• Editor: Monaco Editor (@monaco-editor/react)
• Visualization: Recharts (D3), SVG Progress Rings

Backend

• Core: FastAPI, Python 3.10+
• AI/CV Models: OpenCV, DeepFace, MediaPipe, TensorFlow, PyTorch
• LLM Integration: Groq API (Llama3-8b-8192) for AI hints
• Database: SQLite3 for persistent metrics and adaptive telemetry logs

---

🚀 Local Installation

Prerequisites

1. Python 3.10 or higher
2. Node.js 18+ and npm
3. A Groq API Key (for LLM hints)

1. Clone & Setup Environment

git clone https://github.com/Shridipa/Face-Code.git
cd Face-Code

# Create and activate a virtual environment
python -m venv venv
# On Windows:
.\venv\Scripts\activate
# On Mac/Linux:
source venv/bin/activate

# Install Python dependencies
pip install -r requirements.txt

2. Configure API Keys

Create a .env file in the root directory:

GROQ_API_KEY=your_groq_key_here

3. Start the Backend API

Run the FastAPI server from the root directory:

python main.py
# The local AI models will initialize. The server will host on http://127.0.0.1:8000

4. Start the Frontend Application

Open a new terminal, navigate to the frontend folder:

cd frontend
npm install
npm run dev
# The React application will host on http://localhost:5173

---

🤝 Contributing

Contributions, issues, and feature requests are welcome!

📄 License

This project is proprietary and built for demonstration purposes.