Young Gwak
Software Engineer at Google
Sunnyvale, California, United States
Summary
🤩
Rockstar🎓
Top SchoolYoung Gwak is a software engineer based in Sunnyvale with 8 years of experience specializing in model optimization and TPU performance engineering at Google. He has focused on quantization and inference efficiency—contributing to TensorFlow by improving TensorFlow Lite conversion latency and fixing MLIR quantizer bugs that affect softmax and residual connections. Prior roles blend machine learning and mobile engineering at NAVER, where he built Core ML-based pose estimation iOS apps that measure inference time and visualize heatmaps. Comfortable across research-adjacent systems and production deployment, he brings a practical eye for squeezing latency and accuracy gains from models while shipping mobile and server-side solutions.
8 years of coding experience
5 years of employment as a software developer
학사 컴퓨터 공학(소프트웨어), 학사 컴퓨터 공학(소프트웨어) at Pusan National University
Korean, English
Stackoverflow
168reputation
10kreached
0answers
6questions
Github Skills (21)
quants10
ios10
mobile-development10
python10
machine-learning10
mlr10
coreml10
tensorflow-lite10
tensorflow10
quantization10
swift10
user-interface9
deep-neural-networks9
neural-network9
deep-learning9
Programming languages (10)
Github contributions (5)
tucan9389/PoseEstimation-CoreML
Jul 2018 - Aug 2021
The example project of inferencing Pose Estimation using Core ML
Role in this project:
Mobile Developer (iOS)
Mobile Developer (iOS)Contributions:108 commits, 7 PRs, 93 pushes in 3 years 1 month
Contributions summary:Young appears to be focused on developing an iOS application for pose estimation using Core ML. Their initial commit sets up the basic app structure, followed by commits that integrate the Vision framework for pose detection. The user then added functionalities for measuring inference time and heatmap visualization, and refined the UI components and layout. Further commits involved the creation of a pose-matching feature, indicating a progression toward a more feature-rich mobile application.
ios11deep-learningpose-estimationmachine-learningcoreml
tensorflow/tensorflow
May 2018 - Jun 2018
An Open Source Machine Learning Framework for Everyone
Role in this project:ML Engineer
ML EngineerContributions:4 reviews, 3 commits, 2 PRs in 1 month
Contributions summary:Young contributed to the TensorFlow project by implementing and testing features related to variable quantization, aimed at improving the efficiency of TensorFlow Lite models. They addressed latency issues in Lite conversion by optimizing code, specifically replacing input tensor calls with input tensor values. Furthermore, the user fixed issues in the MLIR quantizer related to softmax operations and addressed bugs related to residual connections in quantized models. These efforts collectively enhance the functionality and performance of TensorFlow's model quantization capabilities.
pythondata-sciencedeep-learningmlmachine-learning
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.