고정 헤더 영역
상세 컨텐츠
본문
Hello, and thank you for visiting my page.
My name is Saebyeok (Dawn) Chu, and I am a South Korean software engineer and aspiring researcher preparing for graduate studies in Computer Science, with a strong interest in software reliability, distributed systems, and AI-driven program analysis. I am currently admitted to the MSc (Thesis) program at Concordia University (Summer 2026) under the supervision of Prof. Yuhong Yan.
My Academic & Professional Journey
I completed my B.S. in Electrical and Computer Engineering at Handong Global University, graduating Summa Cum Laude, with three Dean’s List awards. During my undergraduate years, I also participated in a research internship where I improved a parallel all-pairs shortest path (APSP) algorithm using OpenMP and MPI, analyzing synchronization and memory issues in distributed environments.
From 2021 to 2024, I worked as a Software Engineer at POSCO DX, where I built cloud-based manufacturing automation systems, real-time data pipelines, and anomaly detection tools. This role strengthened my interest in system dependability, large-scale analytics, and data-driven debugging.
I later worked at ShareLife and as a freelance full-stack developer, creating full cloud-native platforms using Django, AWS, Next.js, and PostgreSQL. Through these projects, I learned how to design systems that balance performance, cost, and maintainability—skills I now hope to extend into research.
Current Research: “Bytecode Naturalness”
I am currently working on a research manuscript tentatively titled “Bytecode Naturalness: Statistical Modeling of JVM Bytecode for Bug Detection.”
This work explores:
- whether bytecode, like source code, displays statistical naturalness
- how n-gram models over bytecode sequences can detect anomalous patterns
- how this approach compares to traditional rule-based and source-based detection
- whether bytecode-level modeling offers advantages in language-agnostic or obfuscated environments
My project extends ideas from Bugram, n-gram language models, and software naturalness research, but applies them to compiled code instead of source tokens.
I plan to finish the first polished version around early 2026.
My Research Interests
I am especially drawn to problems at the intersection of software engineering, program analysis, and large-scale systems, including:
- Statistical and ML-based bug detection
- Bytecode-level analysis and cross-language program representations
- Distributed systems optimization and cloud orchestration
- Reliability engineering for real-world long-running systems
- Graph-based analysis of microservice dependencies
- AI-assisted debugging and observability
Across all these topics, I enjoy taking real system problems, understanding their underlying structure, and building practical, scalable solutions.
What Motivates Me
My work at POSCO DX taught me that real-world systems demand reliability, clarity, and thoughtful engineering. My research internships showed me the intellectual excitement of tackling algorithmic and systemic problems. And my recent projects—both academic and personal—gave me confidence that I can contribute meaningful ideas to the research community.
I hope to continue building work that is technically rigorous, grounded in real systems, and useful to both academia and industry.
Beyond Research
Outside of engineering, I enjoy:
- running a small YouTube channel where I teach English and tech topics
- building community through guesthouses, travel, and writing
- studying languages and exploring new cities
- developing side projects that blend creativity and technology
These experiences help me stay curious and open, which influences both my engineering work and my research thinking.
Thank You
If you are a professor or researcher visiting this page, thank you for taking the time to learn about me.
I’m always happy to discuss my interests, share my ongoing work, or explore potential collaborations.
You can reach me at: dawnsaebyeokchu@gmail.com
Also, you can find out my projects at: https://github.com/saebyeokchu/manual
