Overview
LinkNest is a smart link manager that automatically fetches title, description, and images from any URL and categorizes them based on content. It features a clean, modern interface with search, filtering, and favorite functionality, making it easier than ever to organize your digital life.
Technical Implementation
Nuxt 4 Server API Integration
Leveraged Nuxt 4's new server API capabilities to build a robust backend integrated directly with the frontend, handling database operations, authentication, and API routes seamlessly.
Smart Metadata Extraction
Implemented a robust metadata fetching system using Microlink API that handles JavaScript-rendered sites and falls back to basic parsing when necessary, ensuring rich previews for all links.
Optimized Database Architecture
Designed a normalized PostgreSQL schema with proper indexing strategies on the 'links' table to ensure sub-millisecond query performance even as the user's collection grows.
Secure Authentication System
Built a custom JWT-based authentication system with bcrypt password hashing and HTTP-only cookies, ensuring secure session management without relying on third-party auth providers.
Key Features
- Automatic metadata fetching (title, description, images) using Microlink API
- Smart categorization based on content patterns
- Secure JWT authentication with HTTP-only cookies
- Real-time search and filtering capabilities
- Responsive design with dark mode support
- PostgreSQL database with optimized indexing
- Drag-and-drop friendly interface
- Favorites system for quick access
Screenshots
Clean dashboard with automatically categorized links
Smart link addition with automatic metadata fetching
Instant search and filtering by category
Fully responsive design for on-the-go access
Challenges & Solutions
Challenge: Reliably scraping metadata from modern JS-heavy websites
Solution: Initially tried simple HTML parsing but switched to Microlink API to handle dynamic content and JavaScript rendering, with a custom fallback system for reliability.
Challenge: Implementing automatic categorization without complex ML
Solution: Developed a keyword-based pattern matching system that analyzes URL structure and metadata to accurately sort links into categories like 'Dev', 'Blog', 'Video', etc., keeping it lightweight and fast.
Challenge: Managing database connections in a serverless environment
Solution: Implemented connection pooling and proper connection management strategies with NeonDB to handle serverless cold starts and concurrent requests efficiently.
Challenge: Ensuring type safety across full-stack Nuxt application
Solution: Utilized Nuxt's auto-imports and TypeScript integration to share types between server API and client components, ensuring end-to-end type safety.