Sicong Liu

Work Experience

Catalog Quality Team

Optimized previous UI by merging separated webpages into one single page with modals using AngularJS, Node.js, AJAX, HTML5, Bootstrap with unit tests using Jasmine and Karma.
Developed a data management platform for storing and updating task information using Spring Framework, My- Batis, RESTful API with AWS RDS as database.
Developed new features for task management of catalog correction, an inner web tool for merging duplicate items, so that users can automatically see assigned tasks by managers ordered by priorities after login, used by thousands of Amazon catalog quality associates.

Supply Chain Platform Team

Built a supply chain management system by using MVC framework with Flask, MongoDB, MongoEngine, Rab- bitMQ and ElasticSearch, and realized features like real-time replenishment and provider ranking.
Efficiently implemented back-end services based on Microservices structure and RPC framework using Apache Thrift and incorporated RabbitMQ as message queue to decouple back-end microservices.
Developed an optimized simulated annealing algorithm for provider bidding,which reduced 10% purchase cost for the company.

Big Data Platform for Real-time Stock Price

Implemented a high-performance data processing platform using Apache Kafka, Apache Cassandra, Redis and Apache Spark Streaming to analyze stock data.
Optimized payload size using Google Protocol Buffer to improve system throughput by 30%.
Developed a dashboard web app using Node.js, D3.js and created a scalable cloud deployment environment using Docker.

[code]

Smart Zillow Clone

Implemented a real estate search website and value prediction system using Service-oriented Architecture.
Optimized payload size using Google Protocol Buffer to improve system throughput by 30%.
Implemented a distributed real estate web scraping system (integrated with TOR network) together with Zillow API to collect real-time property information using MongoDB and RabbitMQ.
Developed a web service with caching using Node.js/Express, Redis, Bing Map API and Material Design

[code]

Elastic Distributed Databases

Developed an elastic distributed database using MySQL replication and used it to serve as the backend for a multi-tier web application running TPC-W benchmark.
Developed a load balancer to route the browsing queries to master and route the ordering queries to slaves in a round-robin manner.
Guarantee the high availability (HA) of the database in the real-time using backup slaves.
Scale out/in the database server according to service level agreement when real-time workload changes.
Monitor the workload/database/operating system and demonstrate metrics using CanvasJS.

[code]

Real time traffic monitoring system

Designed and developed a real time traffic monitoring system using Java, Spring MVC, Spring Boot, Spring Data, Spring Cloud, Maven, JPA, Hibernate, Tomcat, RabbitMQ, MongoDB, WebSocket, HTML, JavaScript, Bootstrap.
Effectively implemented server side REST APIs such as traffic simulator and traffic updater using Spring Data, Spring Boot and Spring MVC.
Designed and implemented back-end services based on Microservices architecture. Incorporated Netflix Eureka as service registration and discovery.
Persisted data to MongoDB using Spring Data at Data Access Layer and incorporated RabbitMQ as message broker to decouple back-end services.

[code]

Tiny short URL full stack

Designed, developed and tested a REST API web app using the MEAN stack for shorten URL service.
Designed and implemented Web UI using AngularJS and Bootstrap.
Implemented MongoDB as data storage and Used Redis as cache layer to increase the response.
Deployed the Nginx server to guarantee the service of load balancer and reverse proxy.
Applied angular-chart.js to visualize the statistic data of short URL accesses including information like Country, Browser, System, etc.

[code]

Mini Zoggle -- Search Engine

Designed a multi-thread web crawler based on Crawler4j and used it to fetch about 120,000 UC Irvine webpages and stored these data in files and implemented text processing APIs to analyze the files and stored data in MongoDB.
Implemented TF-IDF to build inverted-index model to score webpages and improved search result with PageRank algorithm.
Built search engine with web interface using Python flask framework for user input and output display and evaluated result by NDCG method with over 80% correct.

[code]

New York Traffic Collision Mapping

Showed the collision data graphically by plotting the collisions on a map of New York City where the user can browse/filter the mapped data by time and category.
Designed and implemented the frontend using HTML5, CSS, Bootstrap, jQuery and CartoDB API to visualize the traffic collision.
Implemented query operator to extract data under different filter criteria from Vertica database.
Implemented the servlet in Java to communicate between the filter parameters and CartoDB.

[code]