Skip to content

README.md

Latest commit

 

History

History
118 lines (88 loc) · 3.83 KB

README.md

File metadata and controls

118 lines (88 loc) · 3.83 KB

MPLOY - MAC Job Board

Hey there! 👋

This is a modern, intelligent job board platform that automatically aggregates job listings, providing users with a streamlined experience to search, filter, and discover relevant opportunities. The platform updates daily with fresh listings through our smart AI robots.

Features 🚀

  • Jobs update automatically every single day from various sources (automatically deduplicated)
  • We use AI to help fix, sort and summarise the listings
  • You can filter for exactly what you want (e.g. Big Tech Intern Roles for International students)
  • Works perfectly on phone or laptop
  • Server-side rendering (where possible) with Next.js 15 App Router
  • Multiple filters can be applied at once, including text search
  • Desktop/mobile responsive UI: list/details on desktop, modal on mobile
  • State persists in URL: search, filters, pagination (/jobs?q=dev&location=sydney&page=2)
  • Direct job links supported (/jobs/[id])
  • Parallel data fetching for faster loads
  • Real-time job search with debouncing

Frontend

  • Next.js 15: Utilizing the App Router for server-side rendering and optimized client-side navigation
  • React 19: For building the interactive user interface components
  • TypeScript: Ensuring type safety across the codebase
  • Mantine UI: For consistent, accessible UI components
  • Tailwind CSS: For utility-first styling and responsive design

Key Patterns

  • State Management: Start with props; use custom hooks for reusable logic; Context for global state (e.g., job filters).
  • Data Flow: URL as source of truth for filters/search; debounced API calls; prefetching for pagination.
  • Components: Keep thin (50-150 lines); use layouts for shared UI; mark client components with "use client".
  • Features Used: Intercepting routes for modals; Suspense for loading; Error boundaries; useMemo/useRef for optimization.

Structure

src/
├── app/
│   ├── jobs/          # Main jobs route with filters, listing, details
│   │   ├── [id]/      # Dynamic job page
│   │   └── error.tsx  # Job-specific error handling
│   ├── layout.tsx     # Root layout with providers
│   └── page.tsx       # Home (redirects to /jobs)
├── components/
│   ├── jobs/          # Job cards, lists, details, filters
│   └── layout/        # Nav, logo, search bar
├── context/
│   └── filter/        # Filter state provider
└── lib/               # Utils, theme

Backend

  • Server Actions: Answers search and feedback requests
  • MongoDB: Stores job listings and related metadata
  • GoLang: Powers our web robots (this part is not open source)

Getting Started

Prerequisites

  • Node.js 20+
  • Java 17
  • Go 1.21+
  • Docker & Docker Compose
  • Redis

Local Development

cd frontend
npm install
npm run dev

Environment Variables

MONGODB_URI=
MONGODB_DATABASE=default
MONGODB_COLLECTION=listings

NODE_ENV=development

Development Guidelines

Git Workflow

Branch Structure

  • production - Production branch
  • dev - Development branch
  • Feature branches follow the pattern: 
    <component>/<developer>/<feature-name>
Examples:
- backend/edwn/redis-caching
- frontend/sarah/job-filters

Coding Conventions

  • Use kebab-case for files (e.g., product-card.ts); camelCase for hooks (e.g., useCustomHook).
  • Group related components in feature directories (e.g., components/jobs/filters/).
  • Prioritize server components; use "use client" only when needed.
  • Build with scalability and observability in mind (e.g., structured logging with Pino).

Deployment (Development and Production)

  • We are deployed on Azure Container Apps via Github Actions.
  • GitHub Actions -> builds containers -> Azure Containers Registry -> Azure Container Apps.

License

This project is licensed under the MIT License.