Master of Computer science

University of Illinois Urbana-Champaign, Jan 2024 - May 2025

B.S. Computer Science, Data Science

University of Wisconsin-Madison, Sep 2019 - May 2023

Tharzen - Software Engineer Internship

May 2024 - Aug 2024

• Designed and developed a reversible PHP interpreter based on bidirectional evaluation, leading Test-Driven Development (TDD) to ensure robust feature implementation and comprehensive testing.
• Implemented interpreter functionality using Rust classes and traits to enhance the core logic and maintainability of the reversible PHP interpreter.

Spark Education Group - Software Engineer Internship

Aug 2023 - Dec 2023

• Designed and developed a Python application utilizing longchain embedding, chat completion, and vector store, allowing users to interact with the project codebase effectively.
• Led the deployment of a large language model platform using React, Node.js, Python, FastAPI, and PostgreSQL, integrating new features such as user statistics, top prompts display, and custom masks.
• Drove database and API enhancements, contributing to a 30% increase in weekly active users and achieving a peak of 185 daily active users.

Amazon - Software Engineer Internship

May 2022 - Aug 2022

• Designed and implemented the backend API for the changelog function using Node.js and GraphQL, enhancing user interaction with changelog data based on permissions.
• Created GraphQL schema, field resolver, and record resolver for seamless data mapping and CRUD operations, ensuring high performance and scalability.
• Managed the end-to-end development cycle, from design documents and UI wireframes to code implementation, code review, and deployment.

ClaimTrust – A Propagation-Based Trust Scoring Framework for Retrieval-Augmented Generation (RAG) Systems

Professor: Heng Ji, Aug 2024 - Dec 2024

• Developed a graph-based trust scoring algorithm to evaluate document reliability, improving RAG system response quality by 11.2%.
• Processed 814 articles to extract claims and classify relationships for trust propagation.
• Optimized claim extraction and relationship classification, reducing computational complexity from O(n²) to O(n).

Simulation of the Collaborative Automated Driving Systems

Professor: Bin Ran, Feb 2022 - Dec 2022

• working with the prediction group focusing on trajectory prediction of autonomous driving cars and investigating trajectory prediction methods for the project.
• Generating status info of vehicles from Carla simulator in the format that can be used for future tasks as the data for fitting trajectory prediction models.
• Contributed to the unification of simulator export data format and model input data format.

Offloading pushdown db onto SmartNIC based JBOF device

Professor: Ming Liu , Oct 2021 - Dec 2021

• Customize FlexPushdownDB developed based on C++ hybrid with caching and pushdown capabilities and weighted cache policy from AWS cloud to JBOF device.

Auto parking with Gem e4 Vehicle

algorithm developer

• Designed and implemented image recognition and calibration transposition algorithms using lidar and camera sensors to enhance parking system accuracy.
• Developed and deployed an MPC trajectory tracker and hybrid A* route planning for autonomous vehicle navigation.

Cloud-Based IoT Infrastructure for Vehicle Monitoring

AWS developer

• Designed and deployed scalable IoT infrastructure using AWS IoT Core, AWS GreenGrass, and AWS IoT Analytics to monitor data from a fleet of 120,000 simulated vehicles.
• Implemented AWS Lambda functions to analyze CO2 emissions and return diagnostics, utilizing the publish-subscribe MQTT model for seamless device communication.
• Built data visualization pipelines with AWS Firehose, AWS Glue, and AWS Athena, leveraging AWS Sagemaker Jupyter Notebooks for insights.

Farm data management system

Java programmer

• Designed the system and user interface using JavaFX and CSS.
• Utilizing java packages processing data from csv documents for computing and visualizing.
• Allows users to search data from the database on different categories, displaying the data by farm report, annual report, monthly report, and date Range report with percent of total weight.

Account holders household clustering with associate bank

Python programmer

• Custom distance functions based on address and joint account information using scikit-learn hierarchical clustering, and access data from Snowflake flatform.
• The cost of the model using i7-9700 CPUis exponential 1.21 seconds for 1000 records and 43.09 seconds for 10,000 records, and the highest accuracy of the model is 96.6% with different distance threshold.
• Write a distance clustering function to weight address clustering method, utilize shared account relationship columns in clustering function, and test functionality of the project.

Data research on undergraduate enrollment

Data analysis

• Utilized Python to develop web crawlers for collecting data from online databases.
• Preprocessed data by slicing, reformatting, and implementing a pipeline with StandardScaler transformer.
• Analyzed data using PCA and linear regression methods from the PyTorch library.

Website for Chinese Students and Scholars Association at UW-MADISON

Web develope

• Maintained and updated the website for the Chinese Students and Scholars Association at UW-Madison.
• Implemented new features and functionality based on user requirements.
• Collected and analyzed data from blog and tweet sources. Extracted relevant insights and trends from the data for further analysis.