Soonhwan Kwon
Technical Lead at NAVER Corp
Seoul, South Korea
Summary
👤
Senior🎓
Top SchoolSoonhwan Kwon is a technical lead in Seoul with 11 years of experience building LLM-powered search and multimodal foundation models at NAVER, focused on Korean/Japanese pre-training, document expansion, and relevance labeling. He combines research-grade expertise from Samsung SDS with hands-on engineering leadership to deploy generative AI and vision-language models tailored for shopping and search domains. An active contributor to open-source ML (notably MXNet speech recognition examples), he brings practical end-to-end ML pipeline skills—from data handling and optimizers to model fine-tuning and evaluation. Known for bridging research and product, he accelerates model adoption in production while keeping a strong emphasis on data quality and domain-specific metadata engineering.
11 years of coding experience
13 years of employment as a software developer
학사 전기전자공학, 학사 전기전자공학 at 연세대학교
English, Korean
Github Skills (10)
mxnet10
machine-learning10
speech-recognition10
deep-learning10
python10
modeling9
data-processing9
trainings9
tensorboard8
batch-normalization8
Programming languages (2)
Github contributions (5)
apache/mxnet
Mar 2017 - Jan 2019
Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
Role in this project:
ML Engineer
ML EngineerContributions:17 commits, 29 PRs, 68 comments in 1 year 10 months
Contributions summary:Soonhwan primarily contributed to the speech_recognition example within the MXNet repository. Their work included adding new examples, enhancing existing ones with features like bucketing and batch normalization, and fixing prediction-related bugs. They also addressed issues in the optimizer and data handling components, showcasing a focus on improving the performance and usability of the speech recognition functionalities. The user's contributions demonstrate a focus on the end-to-end machine learning pipeline, encompassing model development, data processing, and evaluation within the context of speech recognition.
pythonschedulerdataflowmutationdata-science
ai-adv-lab/deepspeech.mxnet
Apr 2017 - May 2018
Contributions:71 commits, 6 PRs, 65 pushes in 1 year 1 month
pytorchmxnetdeepspeechdeep-learningbaidu
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.