Summary
AI researcher with expertise in healthcare and NLP, currently pursuing a Ph.D. at KAIST. Developed LabTOP, a unified model for lab test outcome prediction, earning Best Paper Award at CHIL 2025. Published research on clinical large language models and decision transformers. Former AI research intern at AIforPet, specializing in data preprocessing and pose estimation for canine gait analysis. Passionate about applying machine learning to solve real-world healthcare challenges.
Education
KAIST (Korea Advanced Institute of Science and Technology)
Ph.D. in AI
Daejeon, S.Korea
Feb. 2025 -
KAIST (Korea Advanced Institute of Science and Technology)
M.S. in AI
Daejeon, S.Korea
Feb. 2023 - Feb. 2025
Thesis Title: LabTOP - A Unified Model for Lab Test Outcome Prediction on Electronic Health Records
University of California, Berkeley
Visiting Student
Berkeley, CA, USA
Aug. 2018 - Feb. 2022
Taken Courses: Multivariable Calculus, CS101, CS188 (opencourse)
POSTECH (Pohang University of Science and Technology)
B.S. in IT Engineering
Pohang, S.Korea
Feb. 2018 - Feb. 2022
Taken Courses: Calculus I, Applied Linear Algebra, Discrete Math for Computer Science, Probability and Statistics for Engineers, Algorithm, Data Structure, Object Oriented Programming, Machine learning, Artificial Intelligence
Research Publications
[1] LabTOP: A Unified Model for Lab Test Outcome Prediction on Electronic Health Records
Sujeong Im, Jungwoo Oh, and Edward Choi
Best Paper (Applications Practice), Conference on Health, Inference, and Learning (CHIL) 2025
[2] Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes
Sunjun Kweon et al.
Findings of the Association for Computational Linguistics ACL 2024, pp. 5148–5168
[3] Clinical Decision Transformer: Intended Treatment Recommendation through Goal Prompting
Seunghyun Lee, Da Young Lee, Sujeong Im, Nan Hee Kim, and Sung-Min Park
arXiv preprint arXiv:2302.00612 (2023)
Honors & Awards
🏆 Best Paper Award (Applications & Practice)
Conference: 6th Annual Conference on Health, Inference, and Learning (CHIL) 2025
Paper: "LabTOP: A Unified Model for Lab Test Outcome Prediction on Electronic Health Records"
Location: Berkeley, CA, U.S.A
Work Experience
AIforPet
AI Research Intern
Seoul, S.Korea
Jun. 2021 - Sep. 2021
- Implemented dog tracking and background removal techniques using OpenCV to preprocess walking video data
- Applied pose estimation models to canine gait videos and conducted comparative performance evaluation of prior models
- Set up a data annotation environment tailored for tagging canine joint positions
- Prepared a labeling pipeline using DeepLabCut; created tutorials and environment setup guides to streamline team onboarding
- Automated dog cropping from video frames using OpenCV and Darknet to reduce annotation workload
- Converted output annotation files from XML to JSON format to improve compatibility with downstream model training pipelines
- Led pilot tagging sessions and defined consistent annotation guidelines to ensure labeling quality and reproducibility across contributors
- Executed frame-by-frame joint labeling using annotation tools on both video and image datasets
Extracurricular Activities
Edlab, KAIST
Lab Manager
Sungnam, S.Korea
Sep. 2025 -
- Initiated a structured lab seminar series for internal research updates
- Authored the inaugural lab operations manual to streamline the on-boarding process for new members and ensure consistency in research activities
- Structured the lab's hierarchy and optimized resource allocation by assigning clear functional roles to each member
- Standardized lab governance by establishing API usage guidelines and fundamental operating procedures
APGC-Lab (POSTECH entrepreneurship support group)
Manager
Pohang, S.Korea
Feb. 2019 - Dec. 2020
- Led overall operations of APGC-Lab and its community sub-division
- Planned and hosted the startup hackathon "Random Talk Show"
- Encouraged autonomous program planning by members to build a creative community
- Led speaker outreach and overall planning for semester-long entrepreneurship community program
- Hosted and mentored participants during various events and sessions
Educational Outreach Programs in POSTECH
Mentor & Program Staff
Various Locations, S.Korea
Dec. 2018 - Jul. 2021
- Participated as a mentor and organizing staff for POSTECH's Creative Camp for high school students (2018, 2019, 2021)
- Mentored BMC practice sessions and led a large-scale 2-day project during the camp
- Guided a student team at Hwirang Youth Startup Camp on business model development and presentation (Feb. 2021)
- Taught App Inventor and gave a lecture on college life at Gachibaeum Winter Science Camp (Jan. 2020)
Skills
LLM Inference and Serving
Huggingface, VLLM, LiteLLM, Unsloth
Clinical Libraries
MEDS, TIMER
Pretrained Models
GPT-2 (53M)
Fine-tuned Models
Llama-3.1-8B (sft), Llama-3-8B (sft-ppo)
Open Source Models for Inference
Llama-3.1-8B, Qwen3-235B-A22B-Thinking-2507-FP8, Qwen3-Next-80B-A3B-Thinking, Qwen3-Next-80B-A3B-Instruct, Llama-4-Scout-17B-16E-Instruct, Llama-4-Maverick-17B-128E-Instruct-FP8
Programming
Python, C, C++, LaTeX
Languages
Korean, English