Skip to content

noireveil/nexora-learn

Repository files navigation

🧠 Nexora Learn

Next-Era Adaptive Coding Education Platform

Next.js React TypeScript TensorFlow.js Pyodide Status

Nexora Learn is an Offline-First coding education platform utilizing On-Device Machine Learning to personalize learning experiences in real-time. Built to run entirely on the client-side without server dependency, ensuring privacy and accessibility anywhere.


✨ Key Features

🚀 1. Adaptive AI Engine (TensorFlow.js)

Unlike static learning platforms, Nexora adapts to you. The system calculates a User Skill Vector in real-time based on:

  • Accuracy Rate & Hint Usage
  • Time Complexity & Execution Speed
  • Consistency (Streak Impact)

The on-device Neural Network (WebGL backend) predicts the optimal difficulty for your next challenge (Easy/Medium/Hard) to keep you in the "Flow State".

🔒 2. Anti-Cheat & Integrity System

Designed to build real muscle memory, not just copy-pasting skills.

  • Anti-Paste Protection: The editor detects large text insertions and blocks submission, encouraging manual typing.
  • Plagiarism Detection: Uses Levenshtein Distance algorithms to detect code similarity against local history, preventing cheating on shared devices.

⚡ 3. Polyglot Code Execution

Run code securely in the browser using sandboxed Web Workers and WebAssembly:

  • JavaScript: Secure new Function sandbox with constraints.
  • Python: Full Python runtime powered by Pyodide (WASM), capable of running logic, math, and data structure operations offline.
  • Web (HTML/CSS): Real-time DOM rendering with "Space-Insensitive" Regex validation for flexible grading.

🌐 4. True Offline-First Architecture

  • No Server Required: All progress, code history, and gamification data are stored in IndexedDB (via Dexie.js).
  • PWA Support: Installable as a native-like app on Desktop and Mobile.
  • Data Sync: Export/Import your progress via JSON to sync between devices manually.

🏗️ System Architecture

This project employs a Clean Architecture separating UI, Logic, and Data Layers, optimized for React 19 performance.

Layer Technology Description
UI Layer Tailwind CSS + Framer Motion Responsive, animated components with Shadcn/UI patterns.
Logic Layer React Hooks + Web Workers State management and off-main-thread code execution.
Data Layer Dexie.js (IndexedDB) High-performance local database for persistence.
AI Layer TensorFlow.js Client-side inference and model management.

Directory Structure

src/
├── app/         # Next.js 15 App Router (Pages)
├── components/  # Reusable UI Components
├── data/        # Curriculum & Challenge JSONs
├── hooks/       # React Custom Hooks (Controllers)
├── lib/         # Utilities (DB Schema, Gamification Logic)
├── ml/          # Machine Learning Models & Feature Extractor
├── services/    # Business Logic (User, Course, Progress)
└── workers/     # Background Threads (Code Executor)

📦 Getting Started

Ensure you have Node.js (v18+) installed.

1. Clone Repository

git clone https://github.com/noireveil/nexora-learn.git
cd nexora-learn

2. Install Dependencies

npm install

3. Run Development Server

Uses Turbopack for instant Hot Module Replacement (HMR).

npm run dev

4. Open Application

Visit http://localhost:3000.

Note: On the first load, the local database will automatically seed with the latest curriculum data.


📚 Curriculum Paths

Nexora Learn currently offers three comprehensive learning paths with Project-Based Learning:

  1. JavaScript Mastery: From Variables to Functional Programming & Higher Order Functions.
  2. Python for Data Science: Syntax, Structures, Modules, and Data Processing logic.
  3. Web Foundation: HTML5 Semantics, CSS3 Styling, Flexbox/Grid, and Responsive Layouts.

Advanced users are fast-tracked to Level 20 (Guru Badge) and Final Projects via the "Warm Start" algorithm.


🚀 Build & Deploy

Production Build

npm run build

Run Production Server

npm start

About

An offline-first coding education platform featuring on-device machine learning for adaptive challenge generation. Built with Next.js 15, TensorFlow.js, and IndexedDB.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages