Sunghwan Kim
머신 러닝 엔지니어 at ScatterLab
Incheon, South Korea
Summary
👤
Senior🎓
Top SchoolSunghwan Kim is a machine learning engineer with 8 years of experience building and operating production-grade ML systems and LLM serving infrastructure. At ScatterLab he optimized LLM serving for a conversational AI product—cutting serving costs by 45%—and built a multi-cloud Kubernetes operator to autoscale GPUs across seven+ environments. He designs end-to-end MLOps and data pipelines (Airflow, Apache Beam, BigQuery) for training, pseudonymization at nTB scale, and continuous model learning, and has shipped a customer-facing Pingpong AI platform with training-as-a-service capabilities. Previously he developed real-time on-device audio source separation engines (TensorFlow Lite/CoreML/C++) and improved inference speed via quantization and NNAPI. Comfortable across backend (Kotlin/Spring Boot), infra, and model optimization, he blends low-level performance tuning with large-scale orchestration. Outside work he pursues music and audio processing, informing his pragmatic approach to audio-ML problems.
8 years of coding experience
학사, 컴퓨터소프트웨어학부, 학사, 컴퓨터소프트웨어학부 at 한양대학교
Github Skills (42)
rl-environment10
starcraft10
hearthstone10
development-environment10
simulator9
ide9
python-api9
sampler9
vscode9
cpp8
reinforcement-learning8
browser8
javascript7
wave7
vscode-remote7
Programming languages (7)
Github contributions (5)
FYLSunghwan/react-native-piano-sampler
Sep 2020 - Oct 2020
React Native Piano Sampler : Reads sf2 with FluidSynth(on Android) or sfz with AudioKit(on iOS)
Contributions:4 releases, 46 commits, 12 PRs in 1 month
audiokitreactsfzreact-nativeandroid
FYLSunghwan/2048pp
Oct 2019 - Oct 2019
Contributions:62 commits, 38 pushes, 1 branch in 7 days
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.