Xinya Yan

CS & Math Undergraduate student @ UW-Madison

Madison, Wisconsin

LinkedInX
XY

About

I am a third year undergraduate student majoring in Computer Science at the University of Wisconsin-Madison. My previous intern and research experience has been about machine learning, with a focus on neural style transfer and diffusion models. I am currently pursuing research interests in Large Language Models (LLM), especially multimodal models and causality in LLM.

Education

University of Wisconsin-Madison

Sept 2021 - May 2025 (expected)
Bachelor's Degree in Computer Science GPA: 3.978/4.0

High School Affiliated to Renmin University of China

Sept 2018 - June 2021
Cambridge International AS & A Level program

Work Experience

Xiaomi Technology
Internship

July 2023 - Sept 2023

Machine Learning Engineer

Participated in a pioneering image deblurring project, emphasizing the integration of Segment Anything Model (SAM) to address non-uniform image blurring. Focused on optimizing image clarity by blending SAM-generated segmentation masks with image encoding. Utilized the Adam optimizer for model refinement, and rigorously evaluated enhancements using PSNR and SSIM metrics, significantly boosting the performance and accuracy of Xiaomi’s image enhancement technologies.

Foxit Software
Internship

May 2023 - June 2023

C++ Testing Engineer Intern

C++ Unit Testing: Utilized Google Unit Test framework to perform meticulous unit testing on core codes, ensuring the stability and reliability of the software. Developed and executed detailed test cases, focusing on the functionality of PDF file manipulation and character encoding/decoding, resulting in robust and error-free software components.

Skills

Python
Java
C/C++
SQL
PyTorch
TensorFlow
NumPy
Pandas
Spark
Cassandra
Kafka
MySQL
NoSQL
BigQuery

Projects

Cassandra-Parquet Data Integration and gRPC Implementation

Efficiently managed and processed a dataset of 123,346 lines of GHCNd stations data, demonstrating skill in data handling and optimization. Integrated Cassandra databases with Apache Spark to streamline data processing and storage, focusing on efficient data management. Utilized Apache Spark for data conversion to the Parquet format, enhancing data storage efficiency. Implemented gRPC for effective client-server communication, enhancing CQL query processing and data exchange.

Cassandra
Apache Spark
gRPC

Bike Sharing Demand Analysis and Prediction

Analyzed the Kaggle Bike Sharing Demand dataset, focusing on data distribution and key features like time, season, and temperature. Conducted comprehensive data preprocessing, including date handling, missing value treatment, and anomaly detection. Developed predictive models using Support Vector Machine and Random Forest algorithms using scikit-learn

scikit-learn
Support Vector Machine
Random Forest