GenFit AI

GenFit AI is an intelligent, holistic fitness and wellness platform designed to provide highly personalized health and fitness solutions. Leveraging Artificial Intelligence (AI), Machine Learning (ML), and Computer Vision (CV), the platform delivers comprehensive fitness management. Unlike traditional fitness apps with generic solutions, GenFit AI adapts to individual health profiles, medical conditions, and dietary restrictions to promote a sustainable, healthier lifestyle.

• Generates highly customized workout routines using Google Gemini AI.
• Considers constraints like fitness goals, strength levels, medical conditions, and time commitment.
• Provides dynamic and progressive weekly plans.

• Creates personalized dietary recommendations aligned with fitness goals and BMI.
• Factoring in individual allergies, diseases, and food preferences.
• Offers complete macronutrient breakdowns for daily meals.

• Automated food recognition through deep learning (ResNet50 model).
• Users can upload images of their food, and the AI will estimate portion sizes and calculate calories and nutritional value instantly.

• Real-time exercise posture tracking and form correction using TensorFlow (Pose detection) and WebRTC.
• Voice feedback functionality (English & Hindi) for hands-free coaching.
• A responsive FitBot/Agentic Chatbot for answering fitness queries contexts based on user profiles.

• Engaging reward system features points, daily streaks, and milestone badges.
• Leaderboards to motivate users competitively.
• Community connections, social feeds, and support groups.

• Comprehensive user dashboard displaying real-time statistics.
• BMI Calculator with customized AI health insights and tracking.
• Visual charts for monitoring workout adherence and calorie consumption.

Frontend (Client Layer)

• Framework: React.js 18.x with Vite
• Styling: Tailwind CSS, Shadcn UI
• State Management: Redux Toolkit
• Additional Libraries: Framer Motion (Animations), Chart.js & Recharts (Analytics), React WebCam

Backend (Application Layer)

• Environment: Node.js
• Framework: Express.js
• Authentication: JWT, bcryptjs, Google OAuth 2.0
• Real-time Communication: Socket.io

Database (Data Layer)

• DBMS: MongoDB (Atlas)
• ODM: Mongoose

AI & Machine Learning

• Generative AI: Google Gemini API (Groq SDK)
• Computer Vision (Food Image): TensorFlow & pre-trained ResNet50
• Pose Detection: Mediapipe & TensorFlow JS Models

GenFit AI consists of a decoupled architecture with a React Frontend and an Express.js Backend. The system interacts seamlessly with MongoDB for data persistence and third-party AI APIs (Google Gemini, TensorFlow models) to fetch intelligent responses and recognition vectors.

• Node.js (v18+)
• MongoDB Instance (Local or Atlas)
• API Keys (Google Gemini API, Google OAuth, Cloudinary etc.)

1. Clone the repository:

   git clone <repository-url>
   cd MindFit-AI

2. Frontend Setup:

   cd frontend
   npm install
   # Create a .env file based on environment requirements
   npm run dev

3. Backend Setup:

   cd backend
   npm install
   # Create a .env file with necessary variables (MONGO_URI, JWT_SECRET, GROQ_API_KEY, etc.)
   npm start
   # Or for development: npm run dev

(Alternatively, you can run npm run dev from the frontend directory if a concurrent script is configured)

• Log a Workout Session: +10 Points
• Log Food Calories: +5 Points
• Maintain a 7-Day Streak: "Consistency" Badge + Bonus Points
• Check Leaderboard: Climb the ranks every week and complete the Weekly Challenges.

This project is created for educational and development purposes by its respectful authors.