Contact

<aside> <img src="/icons/mail_gray.svg" alt="/icons/mail_gray.svg" width="40px" /> [email protected]

</aside>

<aside> <img src="/icons/phone-call_gray.svg" alt="/icons/phone-call_gray.svg" width="40px" /> +821073278980

</aside>

Skills

🌐 Python

🌐 PySpark

🌐 PyTorch

🌐 MySQL

🌐 R

🌐 JavaScript

🌐 HTML/CSS

🌐 Firebase

…

Education

<aside> <img src="/icons/arrow-right_gray.svg" alt="/icons/arrow-right_gray.svg" width="40px" /> MSc (Computer Science), KAIST Data Science Group CGPA: 4.0/4.3

</aside>

<aside> <img src="/icons/arrow-right_gray.svg" alt="/icons/arrow-right_gray.svg" width="40px" /> BSc (Computer Science), KAIST 152 credits CGPA: 3.76/4.3

</aside>

About Me

I love finding beauty in huge datasets. I am currently working as a contractor on the research team of the Wikimedia Foundation, where I build models to support Wikipedia users.

<aside> <img src="/icons/link_gray.svg" alt="/icons/link_gray.svg" width="40px" /> Webpage

</aside>

</aside>

<aside> 📃 Google Scholar

</aside>

Work Experience

<aside> <img src="/icons/globe_gray.svg" alt="/icons/globe_gray.svg" width="40px" /> Researcher Wikimedia Foundation (Feb 2024 - Present)

Designed multilingual reference need detection system for Wikipedia [link]. Reduced model inference time by a factor of 3.
Designed language-agnostic reference risk system for Wikipedia [link].
Fine-tuned Masked Language Models (MLMs) using huggingface’s transformers.
Retrieved large textual datasets using SQL queries from distributed Hadoop clusters.
Optimized Spark configurations for large-scale data processing tasks. </aside>

<aside> <img src="/icons/activity_gray.svg" alt="/icons/activity_gray.svg" width="40px" /> Student Researcher IBS Data Science Group (Mar 2022 - Feb 2024)

Led a research project analyzing continuous signals from Samsung Watches worn by 50 users over 30 days to predict mental disorders, which earned an award and resulted in an invited journal publication.
Developed an ML-based algorithm to improve customs fraud detection by computing concept drift and applied the algorithm to 4 real-world customs datasets. </aside>

<aside> <img src="/icons/globe_gray.svg" alt="/icons/globe_gray.svg" width="40px" /> Research Intern Wikimedia Foundation (Apr 2022 - Aug 2022)

Designed data pipelines to process large-scale Wikipedia datasets with over a billion raw records, including article texts, edit history, and traffic data using Apache Spark and Python.
Applied advanced statistical modelling techniques such as propensity score matching (PSM) to estimate the effect of user editing experience on referencing of Wikipedia articles. </aside>

<aside> <img src="/icons/code_gray.svg" alt="/icons/code_gray.svg" width="40px" /> Infrastructure Engineer Intern Lucentblock (Jun 2021 - Aug 2021)

Deployed local server for development using Golang, Kubernetes, Pulumi, and Docker.
Designed dashboards in Grafana visualization platform. Implemented alerts from Datadog monitoring tool. </aside>

<aside> <img src="/icons/computer_gray.svg" alt="/icons/computer_gray.svg" width="40px" /> Research Intern KAIST Interaction Lab - KIXLAB (Feb 2020 - Sep 2020)

Developed a web interface with JavaScript/HTML/CSS to collect data via Firebase.
Conducted user studies with 32 participants and analyzed user interaction logs with software tutorial videos. </aside>

Publications

Portfolio Items

Projects

Portfolio Items (1)

Portfolio Items (2)

Invited Talks

“Improving knowledge integrity in Wikimedia projects” Wikimedia Research Showcase, July 2023

This notion template was used on this page.