NutriLife

NutriLife is an app that creates personalized diets according to each user's profile, using our business rules and leveraging information through integrated artificial intelligence.

This repository contains the complete solution, development, technologies, and instructions for running and contributing, should you be interested.

Overview

The project is a modern, robust, scalable, and secure Expo application, divided into two parts:

• Client: Interface developed in TypeScript, using React's componentization patterns and consuming a custom API provided by the server.
• Server: Hosted Node.js API, responsible for the business logic and managing the operation and behaviors of the artificial intelligence.

Both modules follow a modular architecture, recommended standardization practices as mentioned above, centralized configuration, and use up-to-date JavaScript ecosystem tools.

Features

• Onboarding Experience: Welcome screen with interactive carousel showcasing app features
• User Authentication: Complete registration and login system with JWT (JSON Web Tokens)
• User Registration: Allows new users to create accounts with email and password
• Personalized Data Collection: Collects detailed user information:
  • Personal data (name, age, weight, height, gender)
  • Physical activity level (none, light, moderate, heavy)
  • Goal (hypertrophy, definition, weight loss)
• AI-Powered Nutrition Plans: Generates personalized nutrition plans using Google Gemini AI (Gemini 2.5 Flash)
• Nutrition Plan Management:
  • Create personalized nutrition plans
  • List all plans created by the user
  • Detailed view of each plan with meals, macronutrients, and recommendations
• Persistent Storage: Local storage of tokens and user data with AsyncStorage
• Offline-First Architecture: Intelligent request caching with React Query

Technologies Used

Frontend Main libraries:

• React Navigation: Navigation between screens (Stack Navigator)
• Zustand: Global state management (authentication and user data)
• React Query (TanStack Query): API data caching and synchronization
• React Hook Form + Zod: Form validation with TypeScript
• Axios: HTTP client with interceptors for authentication
• AsyncStorage: Persistent local storage
• React Native Swiper: Carousel component for onboarding

Backend Main libraries:

• Fastify: Fast and efficient web framework
• Google Generative AI: Integration with Gemini AI for nutrition plan generation
• Better SQLite3: High-performance SQLite database
• JWT (JSON Web Tokens): Authentication and authorization
• Bcrypt: Password hashing for security
• CORS: Cross-Origin Resource Sharing configuration

For more details about the libraries used, check the package.json files

Client Specifications

• Modular Architecture: Reusable components organized by functionality
• Screen-Based Structure: Screen-based structure (Welcome, Login, Register, Submit, Create, Nutrition, NutritionList)
• State Management:
  • Zustand for global state (authentication and user data)
  • React Query for API data caching and synchronization
• Form Validation: Robust validation with React Hook Form and Zod
• Type Safety: Strict typing with TypeScript throughout the application
• API Integration: Centralized HTTP client with interceptors for automatic authentication
• Navigation: Typed navigation with React Navigation (Stack Navigator)
• Error Handling: Network and authentication error handling
• Loading States: Loading states and visual feedback
• Code Quality: Standardization with ESLint and Prettier
• Configuration: Centralized configuration (env, build, etc.)

API Usage Client Side

The client automatically detects the API base URL depending on the platform:

• Android Emulator: http://10.0.2.2:3333
• iOS Simulator: http://localhost:3333
• Physical Device: Configure the local IP in the src/services/api.ts file

Server Specifications

• RESTful API: REST API structured in controllers and services
• Architecture Pattern: Separation of concerns (Controllers → Services → Database)
• Authentication: JWT middleware for route protection
• Database: SQLite with better-sqlite3 for local storage
• AI Integration: Integration with Google Gemini AI (Gemini 2.5 Flash) for nutrition plan generation
• Error Handling: Centralized error handling with descriptive messages
• Security:
  • Password hashing with bcrypt
  • JWT authentication
  • Input data validation
• Type Safety: Strict typing with TypeScript across all layers
• Code Quality: Rigorous application of lint/format and TypeScript
• Configuration: Isolated environment configuration with .env
• Routes: Organized routes protected by authentication when necessary

API Routes

Public Routes:

• POST /auth/register - Register new user
• POST /auth/login - Login existing user

Protected Routes (require JWT authentication):

• GET /auth/me - Get authenticated user data
• POST /nutrition/create - Create new nutrition plan
• GET /nutrition/list - List all user plans
• GET /nutrition/:planId - Get details of a specific plan

Database Structure

The server uses SQLite with the following tables:

• users: Stores user information (id, email, password, name, createdAt, updatedAt)
• nutrition_plans: Stores created nutrition plans (id, userId, data, createdAt, updatedAt)

The database is created automatically on the first server run.

Environment Configuration

Server (.env)

DATABASE_URL=file:./dev.db
JWT_SECRET=your-jwt-secret-key-here
API_KEY=your-google-gemini-api-key-here

Prerequisites

• Install Node.js (version 18 or higher)
• Install pnpm: npm install -g pnpm
• Android Studio is recommended for emulating the app

Getting Started

1. Clone the repository

git clone https://github.com/zcriticz/nutrilife
cd nutrilife

2. Install dependencies

• Frontend

cd client
pnpm install

• Backend

cd ../server
pnpm install

1. Run the application

• Client Side

cd ../client
pnpm expo start

• Server Side

cd ../server
pnpm dev

⚠️ Important: configure the .env files on both sides (client and server) as needed for local/remote configuration. Without this, the API will not work.

Build

Build Preparation

1. Install EAS CLI

• You can install it globally on your computer:

npm install -g eas-cli

• Or use it locally:

pnpm dlx eas-cli

2. Create Expo Account

• Create an Expo account to use EAS Build. You can do this through the Expo app or website. Then, log in to your account:

eas login

3. Initialize EAS

cd client
eas init

Code Contribution Standards

• Code Style: Maintain the standard defined by ESLint and Prettier (automatic execution on commits)
• Commits: Use Conventional Commits for commit conventions
• Documentation: Add documentation for each new feature (README in the corresponding directory)
• Pull Requests: For bug fixes and features, submit descriptive pull requests
• Issues: Before contributing, open issues to align on major changes
• TypeScript: Maintain strict typing throughout the code
• Testing: Add tests when appropriate (future)

Authors:

• Cristian Santos - Fullstack Developer
  • GitHub Profile
• Mikaias Santos - Designer
  • GitHub Profile

License

This project is licensed under the MIT License. See the license file for more information.