Yongseop Kim
Staff Software Engineer at Samsung Electronics
Seoul, South Korea
Summary
👤
Senior🎓
Top SchoolYongseop Kim is a Staff Software Engineer based in Seoul with 11 years of experience building runtimes, compilers, and on-device AI tooling at Samsung. He has deep hands-on expertise in accelerating neural network runtimes (ONE), developing NPU chip compilers, and extending developer tooling such as a VS Code extension for the ONE compiler toolchain. Yongseop’s background spans low-level system work—from C/C++ native APIs and .NET Core runtime/JIT on Tizen to FP32/FP16 compiler support and CPU backend optimizations—bridging research, performance engineering, and product-focused delivery. An active contributor to high-profile open-source projects like Samsung/ONE and TizenFX, he blends practical performance tuning with maintainable API design. Colleagues describe him as someone who surfaces memory and numeric optimizations others miss, turning subtle efficiencies into measurable runtime gains.
11 years of coding experience
10 years of employment as a software developer
학사 예정, Digital Contents, 학사 예정, Digital Contents at 세종대학교 / Sejong University
English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (15)
asp-net10
net10
dotnet10
c-language10
api10
tizen10
cprogramming-language10
performance-optimization10
csharp10
back-end-development10
dotnet-core10
compiler-design9
xpack9
tensorflow-lite8
build-automation7
Programming languages (4)
Github contributions (5)
Samsung/ONE
Apr 2020 - Jun 2022
On-device Neural Engine
Role in this project:
Back-end Developer & Performance Engineer
Back-end Developer & Performance EngineerContributions:332 reviews, 73 commits, 107 PRs in 2 years 2 months
Contributions summary:Yongseop Kim's contributions primarily revolved around enhancing the "On-device Neural Engine" project. Their work included implementing new functionalities for floating-point operations, specifically introducing conversions between FP32 and FP16 data types to enable FP16 mode in the compiler and related kernels. Additionally, they focused on optimizing performance, as seen in their efforts to release cached constant tensors in the CPU backend, and integrating memory polling to analyze memory usage during model execution. The user also made code improvements for the tools used in the project.
on-device-aion-deviceoptimizationruntimeneural-engine
Samsung/TizenFX
Mar 2016 - Jun 2016
C# Device APIs for Tizen
Role in this project:Back-end Developer
Back-end DeveloperContributions:8 commits in 3 months
Contributions summary:Yongseop focused on enhancing the `AppControl` class within the TizenFX framework. Their contributions included implementing features for manipulating data, such as adding, retrieving, and removing data, and creating an `ExtraData` class. They also implemented `SendLaunchRequest` and `Reply` functionalities. Furthermore, the user refactored `SendLaunchRequest()` and added doxygen comments to the `AppControl` class to improve code readability and maintainability.
dotnetc-sharpcsharptizenfxdevice-apis
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.