Donghyun Lim

Curriculum Vitae

Donghyun LIM (임동현)

• donghyunismyname@gmail.com
• Research Engineer
• Currently working at Nalbi

Skills

C++, CMake, Python, PyTorch, Mathematics, Image Processing, OpenCV

Introduction

Donghyun is currently a research engineer. He works on turning research results into viable products.

He is the type of person who constantly worries about everything, including the software he is in charge of. He hopes his constant worries will eventually lead to reliable products.

He cares about code quality. He likes code that is well-written, beautiful, easy to understand, and easy to maintain. Creating order and regularity out of chaos gives him the utmost satisfaction.

He loves optimization. He has a pathological obsession with making computation faster.

Previously, Donghyun researched theoretical areas of computer science, such as complexity theory or type-2 theory of effectivity, where mathematical properties of computational problems are stated formally and proved rigorously. He hopes his formal math skills learned through theoretical research help develop practical software.

It is not that he ended up in his current position because he wanted it since very young. Life does not go as planned. Life is full of surprises. Donghyun appreciates serendipity. He is willing to learn anything new.

Work Experience

(Apr2022-now) 3D Human Motion Capture with Vision AI at Nalbi

• Algorithm Implementation
  • Devised appropriate geometric algorithms for handling rotations in 3D
  • Optimized image processing pipeline by eliminating existing redundancy in C++ and Swift
  • Developed smoothing algorithms for deep learning output
  • Implemented various mathematical operations in C++
  • Devised an algorithm to measure human motion accuracy score
• ML Inference Pipeline Deployment
  • Implemented full-tracking inference pipeline involving multiple deep learning models in C++ that runs cross-platform: iOS, Android, and Windows
  • Built Windows DLL for ML deployment with C++ and OpenVINO
  • Built iOS xcframework for ML deployment with C++ and CoreML
  • Built Android AAR library for ML deployment with C++ and Tensorflow Lite
• Miscellaneous
  • Encrypted and decrypted deep learning model weight and model output using AES256
  • Processed body pose datasets for deep learning model training
  • Commnunicated as a mediator between machine learning researchers and software developers
  • Helped Unity/Unreal/iOS developers for handling mathematical operations

(Sep2021-Mar2022) Barcode Decoding from RGB Images at Nalbi

• Developed a classical lightweight computer vision algorithm that improved barcode decoding ratio
• Performed experiments on image enhancement using GAN with PyTorch
• Implemented deep learning pre/postprocessing algorithms in C++
• Converted PyTorch models into TensorRT models for on-device deployment

Education

• (2012-2017) B.S. in Computer Science and B.S. in Mathematics (double major), Chungnam National University
• (2017-2019) M.S. in Computer Science, KAIST
  • thesis
  • Complexity and Real Computation Laboratory
• (2019-now) Ph.D. candidate in Computer Science, KAIST, all but dissertation
  • Currently on leave of absence
  • Complexity and Real Computation Laboratory

Publication

• ORCID

Preprints

• https://arxiv.org/abs/2002.04005
• https://arxiv.org/abs/2305.03439

Patents

• (2018) 소셜 네트워크에서 두 사람을 연결하는 근사 최단 경로 검색 시스템 및 방법(Approximate Shortest Path Search System and Methods of Connecting Two People in a Social Network), Patent Registration Number 10-1858178, 대한민국특허청(Korean Intellectual Property Office)

Awards and Honors

• (2018) 5th place, Samsung Collegiate Programming Cup (SCPC) 2018
• (2016) Honorable Mention, ACM-ICPC Asia Daejeon Regional KAIST 4-5th, November 2016
• (2014) Bronze Medal for Non-math Majors, 33rd University Students Contest of Mathematics, Korean Mathematical Society

Internships

• (2018 Jul-Aug) Internship, Software Engineering, Elice
  • Customized Jupyter Notebook to capture the interaction between frontend and the language kernel
• (2018 Jan-Feb) Internship, Software Engineering, Elice
  • Expanded source code clone detector to support not only Python, but various other languages such as C, C++, and Java
  • Program syntax analysis
• (2017 Jan-Feb) Chungnam National University / Purdue University National Center for Excellence in Software Program 2017
  • Worked on a project for acoustic UAV detection. Applied convolutional neural networks to recognize sound fingerprints generated by UAVs.
• (2016 Jul-Aug) Internship, Software Engineering and Application Laboratory at Chungnam National University
  • Developed graph visualizer using Python and Graphviz for static program analysis

Languages

• Korean (native)
• English (professional working proficiency)
• Japanese (JLPT N1, highest level)

last update Mar 2024