Cheng-Yi Tang

Cheng-Yi Tang

唐成儀 (Thomas) | Software Engineer & Graduate Student

About Me

I'm a Master of Software Engineering student at UC Irvine with a passion for building scalable software solutions. I have experience in full-stack development, machine learning, and cloud technologies. Currently working at AWS as SDE intern.

Education

University of California, Irvine

Master of Software Engineering

2024 - 2025 | Irvine, CA

National Chengchi University

Bachelor of Science in Management Information Systems

2019 - 2023 | Taipei, Taiwan

Experience

Software Development Engineer Intern

Amazon Web Services (AWS)

Summer 2025 | Seattle, WA

Built scalable ETL pipeline with PySpark and Glue to consolidate multiple data sources for near real-time monitoring on AWS QuickSight.

Improved AWS Athena query performance by implementing optimized S3 partitioning and query tuning techniques.

Software Engineer Intern

Raydium Semiconductor

Summer 2024 | Hsinchu, Taiwan

Developed a lightweight depthwise CNN (0.06MB, 14K params, 18.46M MACs) for touchscreen environment classification, achieving 94% accuracy on 71K samples using PyTorch.

Optimized model with 2D-to-1D reshape and pruning, reducing parameters by 17% and MACs by 52% with only 1% accuracy loss, enhancing hardware deployment feasibility.

Technical Sales Specialist Intern

Intel

2022 - 2023 | Taipei, Taiwan

Created Python Dash applications for Datacenter and AI design-win project tracking and automated data analytics, reducing reporting time from 4 hours to 1 second.

Introduced High-Bandwidth Memory (HBM) to the media and professionals on Intel Innovation, demonstraing related benchmarks of 4th generation Xeon Scalable processors

Software Projects

Versus

Next.js, FastAPI, Docker, PostgreSQL

A full-stack platform for tracking 1:1 games and sports results with win-rate analytics and league features.

View on GitHub

Apache Cassandra CQL Test Suite Enhancement

Java, JUnit, JaCoCo

Enhanced testing framework for Apache Cassandra, improving code coverage from 29% to 60% with automated CI workflows.

View on GitHub

Sitting Posture Recognition System

Python, OpenCV, MediaPipe, TensorFlow/Keras

A real-time sitting posture detection system using computer vision and deep learning. Achieves 97.78% accuracy in posture classification with VGGNet16 architecture and provides audio alerts for poor posture.

View on GitHub

Writing

I occasionally write about my software engineering learning experiences and life on Medium.

Summer 2025 Intern Summary

從退伍後的 5 月開始投遞實習申請,直到 3 月順利上岸,這段找工作的旅程中受到許多人的幫助,所以想寫篇總結表達感謝。同時也記錄下 2025 年的求職狀況,希望能幫到未來幾年想在美國找工的學弟妹們。

March 2025

Running Official Llama 3.2 on M1 Mac

From model download to local deployment: Setting up Meta's official release with llama.cpp and Hugging Face convert tool.

October 2024

Hacking OWASP Juice Shop—Beginner Experience

This article details my hands-on lab experience with OWASP Juice Shop, an intentionally vulnerable web application designed for security training.

March 2025

Skills

Languages

Java, Python, JavaScript, TypeScript, R, SQL

Web Development

React, Node.js, FastAPI, Django, Flask, HTML, CSS

ML & Data Science

PyTorch, TensorFlow, NumPy, Pandas, Scikit-learn, OpenCV, Keras

DevOps & Databases

Git, Docker, CI/CD, AWS, PostgreSQL, MySQL, MongoDB, DBeaver, Linux, Unix Shell

Beyond Work

When I'm not coding or studying, I enjoy: