FlowCV Logo
About Me
I am passionate about researching cutting edge AI technologies (Diffusion, Vision Transformers, Reinforcement Learning, etc.) and applying them to real-world problems. I have experience in the full ML pipeline — from model design to validation, publication, and implementation — across diverse domains such as robotic control, image generation, and financial forecasting. I place great value on solving practical problems through step-by-step reasoning, grounded in both data and domain understanding. I aim to become a highly capable Software Engineer who bridges theory and application.
Work Experience
Full-time Software Engineer, Dalpha
  • •Implemented VIT-based eye image classification and vector search-based lens recommendation system (PPB Studios).
  • •Developed BERT-based long-text classification model and custom evaluation metrics (Gabia CNS, KT Commerce).
    • 09/2023 – 12/2023
  • •Built a multimodal retrieval prototype using CLIP (for Coding platform).
  • •Worked on full stack development using PyTorch, FastAPI, LangChain, and vector databases.
    • Internship, Kakao Brain
  • •Designed and deployed a custom TTS backend system using Node.js, MongoDB, and GCP.
  • •Built a cloud architecture using Pub/Sub and Compute Engine and collaborated in an agile team.
    • 12/2021 – 02/2022
    Education
    M.S. in Computer Science and Engineering, Seoul National University

    Advisor: Prof. Byungro Moon (Optimization and Financial Engineering Lab)

    03/2024 – 02/2026
    B.S. in Electrical and Computer Engineering, Seoul National University
    03/2017 – 02/2024
    Daegu Science High School
    03/2014 – 02/2017
    Skills
    Languages

    Korean (native), English (fluent in academic writing and reading)

    Programming

    Python, C++, JavaScript, PyTorch, Node.js, React, Git, Docker, Linux, GCP

    Competitions & Awards
    SNU FastMRI Challenge 2022 - Special Prize (Ranked 6th, W1.25M prize)

    Enhanced E2EVarNet with DIRCN-based model to improve MRI reconstruction.

    2022
    Samsung SCPC 2021 - Advanced to Round 2
    2021
    Hyundai MOBIS Programming Contest 2021 - Top 100
    2021
    Research Publications
    Research Internships
    Robotics and Control Lab, Seoul National University (Prof. Songhwe Oh)
  • •Designed a reinforcement learning algorithm for object rearrangement using Signed Distance Field (SDF)-based graph state representation and CCN-based Q-Network.
  • •Improved object correspondence in scene graphs to reduce matching errors; leveraged Hindsight Experience Replay (HER) to boost sample efficiency.
    • 03/2023 – 06/2023
  • •Integrated CPD-based rigid point set registration and SuperPoint-based visual matching.
    • Computer Vision Lab, Seoul National University (Prof. Bohyung Han)
  • •Implemented YOLO, DETR, and Vision Transformer-based detection models; analyzed Deformable- DETR, and DN-DETR.
  • •Studied limitations of transformer-based aggregation in tiny object detection; evaluated multi-scale attention strategies.
    • 06/2022 – 12/2022
  • •Conducted backbone tuning and feature map visualization for robustness analysis.