If cat cries nya, I would cry nye.
Hello, welcome to my page :3


  • My interests lay on lowering the barriers of knowledge via technology and my skills.
  • I love learning, dreaming, contributing to opensource.
  • I’m a dedicated note taker.
  • I name everything.

Contact me on e-mail : iam@nyanye.com

Nyanye (as of September. 2018)

Versatile and flexible Data Scientist, DEVOPS engineer and open source developer.

Work Experience

Full-time Data Scientist, Neurophet, Seoul, South Korea (2017.09 ~)

  • Developed and maintained internal MRI viewer for qualitative evaluation of automated segmentation algorithms
  • Built and maintained google cloud based automated scalable & managed segmentation system
  • Developing generally adaptable fully automated whole-head segmentation application
  • Developing and maintaining internal python library for machine learning, neuroimaging, voxel image processing

Data Scientist part-time Intern, Neurophet, Seoul, South Korea (2017.02 ~ 2017.09)

  • Trained several neural network architectures for medical imaging task, Implemented pre-trained model into tensorflow C++ API


Suwon Hi-tech Highschool

Data Science Self-training, Github ★ 42+


  • Multi dimensional, multi modal medical image processing & learning
  • Python (Tensorflow, Pandas, Keras, Scikit-learn, Nibabel, Numpy, vtk, PyQt, …)
  • Cloud services / softwares operations. (GCP, AWS, docker, linux, ansible, airflow …)
  • Natural Language Processing
  • Tools that I use

Mindmap of technologies that I used so far

Awards & Honor

Machine Learning

2017 Gyeong-gi Province Skills Competitions, 1st Place

  • Classified demographic information by internet history using raw user search queries & statistical Features. Received an award of $1,300

Kakao Arena Shopping Category Classification, 5th Place out of 446 teams

  • Classified hierarchical online shopping category from large datasets including product informations. Received an award of $3,000


MICCAI 2018 - Grand Challenge on MR Brain Segmentation 2018

  • The purpose of this challenge is to directly compare methods for segmentation of gray matter, white matter, cerebrospinal fluid, and other structures on 3T MRI scans of the brain, and to assess the effect of (large) pathologies on segmentation and volumetry.



  • Professional working proficiency


  • bilingual proficiency, Japanese Language Proficiency Test N1 Qualification


  • Native

Projects / Open Source

Koshort project owner (Github ★ 38+, 9+, 6+)

Koshort is a high-level API project written in Python for Korean natural language processing (NLP)

  • Koshort - Korean natural language streaming or crawling library
  • Pyeunjeon - Korean morphological analysis library
  • Goorm - Korean morphological analysis based worldcloud generator

KoNLPy project maintainer (Github ★ 661+)

KoNLPy is an open source project to lower barriers to Korean Natural Language Processing written in Python. Project was originally created by Lucy Park and invited as a project maintainer.

Tweepy contributor (Github ★ 5236+)

Twitter for Python!

hangulize contributor (Github ★ 157+)

Hangulize - (Foreign language to Korean transliterator written in Python)

GPUtil contributor (Github ★ 172+)

GPUtil is a Python module for getting the GPU status from NVIDA GPUs using nvidia-smi. GPUtil locates all GPUs on the computer, determines their availablity and returns a ordered list of available GPUs. Availablity is based upon the current memory consumption and load of each GPU. The module is written with GPU selection for Deep Learning in mind, but it is not task/library specific and it can be applied to any task, where it may be useful to identify available GPUs.