cv

Postbaccalaureate Researcher

Graduate School of AI, KAIST

Worked on foundational model for clinical time-series data. Improved a CycleGAN-based scatter correction software for Chest X-ray images.

• TimeVQVAE
• Wavelet-CycleGAN

Research Intern

DeepMetrics

Involved in a project devising a method for automated control of mechanical ventilators in intensive care units.

• Decision Tree
• Time-Series Data

Research Intern

Seoul National University

Involved in development of Foldseek. Developed optimized code for LDDT alignment score computation, and established an automated protein database update pipeline for Foldseek's web server.

• Spatial Hashing
• SIMD Instructions

M.S.

Korea Advanced Institute of Science and Technology, Seoul, Republic of Korea

Graduate School of AI

B.S.

Seoul National University, Seoul, Republic of Korea

Electrical and Computer Engineering

Summa Cum Laude

Seoul National University

Presidential Science Scholarship

Korea Scholarship Foundation

Full tuition plus 2.5 million won per semester, awarded to talented students majoring in science or engineering.

Medical AI Scholarship

Korea Health Industry Development Institute

3 million won per semester, merit-based scholarship

PCPO: Proportionate Credit Policy Optimization for Aligning Image Generation Models

ICLR 2026

PCPO is an improvement on GRPO for fine-tuning generative models.

Fast and accurate protein structure search with Foldseek

Nature Biotechnology

As structure prediction methods are generating millions of publicly available protein structures, searching these databases is becoming a bottleneck. Foldseek aligns the structure of a query protein against a database by describing tertiary amino acid interactions within proteins as sequences over a structural alphabet. Foldseek decreases computation times by four to five orders of magnitude with 86%, 88% and 133% of the sensitivities of Dali, TM-align and CE, respectively.