Job+ : AWS based Web Service Development – Job Recommendation
Front End:
• Developed an interactive web page for users to search and apply positions (HTML, CSS, JavaScript, AJAX)
• Used favorite records to provide personalized position recommendation
Back End:
• Created three Java servlets with RESTful APIs to handle HTTP requests and responses
• Used MySQL database on Amazon RDS to store position data fetched from Github API
• Used MonkeyLearn API to extract keywords from description of positions
• Designed algorithms (e.g., content-based recommendation) to implement job recommendation
• Deployed to Amazon EC2 for more visibility.
OnlineShop: a Spring and Hibernate based Shopping and Ordering system
Used Spring framework to build a web application for users to shop and order items online.
• Built a web application based on Spring MVC to support item search and listing (dependency injection, inversion of control, REST API etc.).
• Implemented security workflow via in-memory and JDBC authentication provided by Spring Security.
• Utilized Hibernate to provide better support of database operations
• Developed a Spring Web Flow to support item ordering.
Starlink: React JS based Starlink Trajectory Visualization
• Designed and developed a visualization dashboard using ReactJS and D3 to track satellites in real-time based on geo-location.
• Built location, altitude, and duration based selector to refine satellite search.
• Animated selected satellite paths on a world map using D3 to improve the user friendness.
• Deployed the dashboard to Amazon Web Service for demonstration.
Tinnews: a Tinder-like News App
• Designed a Tinder Flavored News app based on Android Architectural Component MVVM Pattern to decouple the components
• Implemented the bottom navigation of three fragments with BottomNavigationView
• Implemented network APIs with Retrofit and LiveData to pull the latest news data from a RESTFUL endpoint (newsapi.org)
• Page navigation with JetPack navigation component, adopted SafeArg for type safety when passing parameters.
• Integrated a 3rd party CardStackView(RecyclerView) to support swipe gestures for like/unlike
• Built the Room Database with LiveData & ViewModel to support local cache and offline model
