Hi, My Name Is
I am Jin-Dong (Mark) Dong, a PhD student in CyLab at Carnegie Mellon University advised by Prof. Nicolas Christin. My research focuses on understanding and protecting users in digital environments through user behavior analytics. By analyzing patterns in how users interact with the systems, I can identify vulnerabilities and predict threats across different domains, such as assessing financial risks in cryptocurrency applications and protecting users from malicious actors online.
Before joining CMU, I received my bachelor’s degree in Computer Science at National Tsing-Hua University. I did research on Neural Architecture Search (NAS) with Prof. Min Sun where we proposed a framework to automatically generate neural networks based on user demands.
Outside of work, I enjoy playing baseball, cooking (especially pasta dishes like the four pastas of Rome), and making tea-like coffee. My name in traditional Chinese (zh-TW) is 董晉東
Selected Projects
Blockchain Fraud Detection & Risk Assessment
Phishing Attack Detection, Dashboard
- Developed and deployed a real-time detection system discovering cryptocurrency address poisoning phishing attacks amounting to more than $1M USD in weekly losses. Built high-performance analysis pipelines capable of processing Ethereum (∼ 12s / block) and Binance Smart Chain (∼ 3s / block) transactions in real time and achieved state-of-the-art detection accuracy.
- Media: Cointelegraph
- Publication: USENIX Security 2025, SBC 2025
Derivative Market Analysis
- Applied blockchain forensics and clustering techniques to analyze amateur v.s., professional trading behaviors on BitMEX, one of the pioneering cryptocurrency derivative exchanges. Supported the understanding of how derivative trading impacted cryptocurrency ecosystem and sparked regulatory discussions.
- Media: The New York Times, Wall Street Journal, The Economist
- Publication: WWW 2021
Machine Learning for Network Security & Privacy
De-anonymization of Tor Onion Traffics
- Analyzed the limitations of industry-standard Machine Learning approaches for correlating Tor onion traffics and helped the design of a subset-sum algorithm based solution.
- Award: Best Portuguese Internet Research of Internet Society 2024
- Publication: NDSS 2024
Education
Carnegie Mellon University
- Ph.D. in Electrical & Computer Engineering
- Sep 2019 - Present
- Pittsburgh, PA
Carnegie Mellon University
- M.Sc. in Electrical & Computer Engineering
- Stemmed from the PhD program, awarded Fall 2023
- Pittsburgh, PA
National Tsing-Hua University
- B.Sc. in Computer Science
- Sep 2013 - Jun 2017
- Hsinchu, Taiwan