Mufeng Xie

979-***-**** adx7po@r.postjobfree.com College Station, Texas

Education

Texas A&M University College Station, TX

Master of Computer Science Sep. 2019 - Aug. 2023

University of Science and Technology of China Hefei, Anhui, China Bachelor of Science in Computer Science and Physics Sep. 2015 - May. 2019 Programming Skills

Programming Languages: C, C++, Python, Java, JavaScript, SQL, HTML, CSS Frameworks: Flask, Spring Boot, Spring Cloud, Node.js, Vue.js, React.js, OpenMP, MPI, ArcGIS Pro, MyBatis Databases: MySQL, MariaDB, Redis, MongoDB, ElasticSearch, SnowFlake, AWS S3, Amazon DynamoDB Software Tools: Git, Linux, Postman, Insomnia, Docker, Bash, SSH, tmux, AWS, JMeter, RocketMQ, Sentinel, Kibana Work Experience

TikTok (Bytedance) Mountain View, CA

Software Engineer Intern, Data-System-Networking Team (SDN Team) Jun. 2022 – Aug. 2022

● Developed a full stack troubleshooting application for network architecture workflows utilizing Vue.js and Flask, and implemented backend features such as extracting workflow structures for graph display using regular expressions.

● Utilized Lark bot API on Docker container to send users troubleshooting alerts, and designed a workflow for testing connectivity of multiple links by utilizing syslog data on Kibana with ElasticSearch.

● Implemented Multi-link-down workflow and constructed the dedupe configuration from state data of links and switches in MySQL database to optimize the application for scalability and performance.

● Built mutiple user-friendly frontend web pages for event triggering, monitoring, updating, and displaying Workflow Checkpoints with VIS.js, and incorporated troubleshooting alert cards directly on the frontend platform (NIB dashboard) for convenient access to troubleshooting information. Academic Projects

Veoride Route Prediction and Distribution Rebalancing Optimization Feb. 2022 – Feb. 2023

● Developed a Veoride route density prediction model utilizing Machine Learning algorithms such as Random Forest and Decision Tree Regression, utilizing Veoride bike route data and log data around college station.

● Implemented an algorithm to optimize the distribution and rebalancing of bikes in real-time based on the predicted usage data and current bike availability to improve operational efficiency and customer satisfaction.

● Built a rebalancing map to optimize cost with ArcGIS Pro, incorporating daily log data and predicted usage data. Competitive Ticket Sales System using SpringBoot May. 2023 – July. 2023

● Built a Competitive Ticket Sale System using Java SpringBoot as framework, mySQL to process ticket sales information, and Redis to improve database I/O for ticket availability with less response times and smoother user experiences.

● Employed MyBatis reverse generation techniques to streamline the integration between Spring application and mySQL.

● Implemented traffic peak clipping techniques to manage spikes in demand leveraging RocketMQ, and simulated and tested the system’s performance using JMeter during high concurrency.

● Ensure the generation of unique and distributed IDs for data integrity and consistency using SnowFlake Algorithm. Article question entailment using Natural Language Processing Feb. 2021 – May. 2021

● Implemented an Article Question Entailment model using Natural Language Processing techniques, utilizing a Boolean Question Answering model from the BoolQ dataset collected by Google Search Engine.

● Utilized BERT and RoBERTa pretrained models and ‘robert-base’ to generate tokenizers for improved accuracy.

● Achieved a highest accuracy of 80.25% on the dev set among all the epoch numbers, which demonstrate the efficiency and effectiveness of the proposed model.

Graduate Student Tracking System following Agile approach Sep. 2020 – Dec. 2020

● Designed and developed a full-stack web application for graduate student tracking using an Agile approach, utilizing technologies such as JavaScript and Google AppsScript.

● Implemented key features such as student searching, displaying student information, adding and editing, and switching between former and current student charts, all utilizing data pre-stored in Google Drive.

Contact this candidate