Education

Sep. 2019 - Aug. 2021

Master of Science

Korea Advanced Institute of Science and Technology (KAIST)

  • Major: Artificial Intelligence
  • Advisor: Prof. Kijung Shin ( Data Mining Lab )
  • Thesis: Continuous CP Decomposition of Sparse Tensor Streams
  • Government-sponsored scholarship

Mar. 2014 - Feb. 2019

Bachelor of Science

Korea Advanced Institute of Science and Technology (KAIST)

  • Major: Electrical Engineering
  • Triple degree 1: B.S. in Computer Science and Engineering
  • Triple degree 2: B.S. in Mathematical Science

Mar. 2012 - Feb. 2014

High School Diploma

Gyeonggibuk Science High School

Working Experience

Nov. 2021 - Current

KRAFTON, Inc. ( KRAFTON AI )

Deep Learning Research Engineer

May. 2026 - Current

AI Applied Research Department

  • Research agentic system for end-to-end game creation

Mar. 2024 - Apr. 2026

Deep Learning Research Department

  • Research anti-cheat system : Video-based ESP/Aimbot detection, ESP cheater simulation
  • Research LLM : Agentic system for playing video games, LLM post-training, LLM pruning and distillation
  • Research speechLLM (Government's sovereign AI foundation model project; Dokpamo) : Transformers/Megatron-based speechLLM training, Evaluation system construction

Nov. 2021 - Feb. 2024

Deep Learning Team | Vision and Animation Part

  • Research stylized face mesh generation : Stylized face recognition, Rigged mesh reconstruction from single-view image, Audio-driven 3D speech animation
  • Research text-to-image diffusion model : Game asset generation, LoRA fine tuning, Image generation service deployment

Nov. 2017 - Feb. 2018

Information Theory and Machine Learning Lab @ KAIST

Research Intern

  • Advisor: Prof. Sae-Young Chung
  • Research reinforcement algorithms to improve the performance of game agents using DQN.

Mar. 2017 - Aug. 2017

Stratio, Inc. (US office)

Software Engineer Intern

  • Global ICT Internship Program (supporter: IITP, FKII)
  • System architecture design and implementation for a spectroscopy-based drug identification product
  • Training pipeline automation : Jenkins-based training queue under limited GPU resources, Django web GUI and model version management for real-time monitoring
  • Model deployment optimization : Server-to-iOS-app model delivery via custom API and protocol, reducing deployment time from 30 minutes to 1 minute by eliminating app rebuilds

Activities

Nov. 2024 - Current

KRAFTON Orchestra

Violinist

  • Performed at the KDC 2025 (Krafton Developer Conference) game music concert

Teaching Experience

Teaching Assistant @ KAIST

  • AI506: Data Mining and Search ( Spring 2020 )
  • AI607: Graph Mining and Social Network Analysis ( Fall 2019, Fall 2020 )
  • AI611: Deep Reinforcement Learning (Spring 2021)

Teaching Assistant @ Samsung

  • Samsung DS AI (Natural Language Processing) Lecture ( Sep. 2019 )
  • Samsung DS - KAIST AI Expert Program (Jul. 2020)
  • Samsung Electronics AI Specialist (Sep. 2020)

Tutor

  • KAIST Undergraduate Research (Winter 2020, Spring 2021)

Copyright 2026. yunik1004 All rights reserved.