Smart Todo Coding Challenge

Build a Smart Todo app with authentication and AI-powered task suggestions.

Rules

1. Screen-record the entire coding process - no cuts, no edits
2. Narrate as you code - explain your decisions and thought process (in English/Russian/Uzbek)
3. Allowed: ChatGPT, Google, Stack Overflow, documentation, copy-pasting snippets, asking AI questions, IDE autocomplete/inline suggestions (Copilot, Codeium, etc.)
4. NOT allowed: Running AI agents in autonomous mode that build the app in one shot (e.g., Cursor Agent, Claude Code in agent mode, Devin, etc.)

Challenge Requirements

Backend (Node.js + Express + MongoDB)

• Register endpoint (POST /api/auth/register)
• Login endpoint (POST /api/auth/login) - returns JWT
• Create todo (POST /api/todos) - protected
• List todos (GET /api/todos) - protected, user's own todos only
• Delete todo (DELETE /api/todos/:id) - protected
• Toggle todo (PATCH /api/todos/:id) - protected
• AI suggestions (POST /api/ai/suggest) - takes a general task, returns 5-7 action items

Frontend (React)

• Register page
• Login page
• Dashboard (protected - only accessible when logged in)
• Input field to add a todo
• "Suggest" button - calls AI endpoint and displays suggestions
• Add suggestions to todo list with one click
• List of todos
• Mark todo as done/undone
• Delete todo

Authorization

• Each user can only see and manage their own todos
• Backend must enforce proper authorization checks

Getting Started

Prerequisites

• Node.js 18+
• MongoDB (local or Atlas)
• OpenAI API key (or any LLM API)

Backend Setup

cd backend
cp .env.example .env
# Edit .env with your MongoDB URI, JWT secret, and OpenAI API key
npm install
npm run dev

Frontend Setup

cd frontend
cp .env.example .env
npm install
npm run dev

Submit

• GitHub repo link
• Unedited screen recording (upload to YouTube, Loom, Google Drive, etc.)