ML Engineer & Data Scientist
I build AI systems that turn data into real-world impact — currently a Decision Intelligence Platform that helps people and businesses make better decisions.
I'm a Machine Learning Engineer and Data Scientist who solves real-world problems with data. I'm currently building a Decision Intelligence Platform at Ailys, helping people and businesses make better decisions and capture more value from them. Before that, I built AI-driven recommendation and knowledge-tracing systems at TmaxEduAI, used by millions of learners, and developed NovoRank, a machine learning system that improves peptide identification accuracy, published in the Journal of Proteome Research, during my M.S. in AI at Hanyang University.
NovoRank: Refinement for De Novo Peptide Sequencing Based on Spectral Clustering and Deep Learning
Jangho Seo, Seunghyuk Choi, Eunok Paek — J. Proteome Res. 2025, 24, 2, 903–910
Read paper →Professional
Bioinformatics · Deep Learning · Clustering
Conventional de novo peptide sequencing tools score spectra in isolation and frequently misidentify peptides. I built a two-step clustering and deep learning re-ranking pipeline that cross-checks candidates against similar spectra, improving precision and recall by an average of 4.6% and 4.5% across three state-of-the-art tools.
Recommendation Systems · Knowledge Tracing
Course recommendations for a 2M+ user platform were generic and didn't reflect learner progress. I built a deep learning recommendation model on sequential interaction data and shipped it through an Airflow pipeline, lifting course enrollment (CVR) by 5% and content completion by 25%.
Computer Vision · Object Detection
Indoor construction sites have small- and medium-sized hazards that are easy for inspectors to miss. As part of an automated safety-rule-checking project, I trained image classification and object detection models to flag these hazards from site imagery.
Personal
Machine Learning · LightGBM
Wanted to understand how far a tuned gradient-boosting baseline could go against a 40M-row, heavily imbalanced ad-click dataset before reaching for deep learning. I built a LightGBM pipeline addressing class imbalance and high-cardinality categorical encoding on the Avazu CTR dataset.
Python, SQL, Java
PyTorch, LightGBM, Airflow
Recommendation Systems, Knowledge Tracing, Bioinformatics, Computer Vision