Juhyun Bae

Juhyun Bae

CS Ph.D. student

Office : KACB 3319
266 Ferst Dr NW, Atlanta, GA 30332

Phone : 770-756-7640

Email : juhyun.bae@gatech.edu




Systems for ML and AI

ML Algorithms for Performance Optimization of Cloud and Cluster Computing Systems

ML and Data Analytics Algorithms for Big Data Applications and Services



Juhyun Bae is a Ph.D student in Computer Science at Georgia Institute of Technology. His advisor is Prof. Ling Liu

and now he is working with Machine Learning, Big Data Analytics and Cloud computing. Prior to this, he did Master’s

in Computer Science at Georgia Institute of Technology for a year and received Bachelor’s in Computer Engineering at

Hanyang University in South Korea.



Cum Laude, Hanyang University                                                                                                                                        Feb 2010

Honor Scholarship for Excellent Academic records, Hanyang University        Mar 2008, Sep 2007, Mar 2007, Sep 2006

Early Identified Pool (EIP), LG Electronics                                                                                  Mar 2013, Mar 2014, Mar 2015



External Reviewer for UCC 2018, “Machine Learning Approach for Live Migration Cost Prediction in VMware




Georgia Institute of Technology                                                                                                                       Atlanta, USA

Graduate Research Assistant                                                                                                                                Aug 2018 – Present

Advisor : Prof. Ling Liu


Nissan Research Center – Silicon Valley                                                                                                  California, USA

Summer Research Intern                                                                                                                                  May 2018 – Aug 2018

Supervisor : Dr. Vikram Krishnamurthy

  • Usage Based Insurance based on Clustering and Machine Learning techniques

          UBI is ‘pay how you drive (PHYD)’, a type of vehicle insurance whereby the cost is dependent upon type of vehicle

          used, driving time, distance, behavior and place. Main goal of this project is to provide rare event (Harsh Accelera-

          tion and Harsh Breaking as a dangerous behavior) detection method through machine learning techniques with

          the massive data from the connected vehicles. Extracting behavioral driving patterns and defining the behavioral

          patterns in various perspectives with clustering.




Implementation of Distributed Key-Value Store System                                                          Apr 2018 – May 2018

Proposed new design of distributed key-value store.Building hierarchical layered consistent hash ring based distributed

key-value store Library

Gesture Detection & Classification                                                                                                    Mar 2018 – May 2018

Main goal of this project was to extract and analyze human behavior through the video data. Creating new feature that

is not linearly corelated to other features and reinforces the characteristics of time series data through KDD(knowledge

discovery in database) 

Implementation of Recoverable Virtual Memory Library                                                         Mar 2018 – Apr 2018

Building Recoverable Virtual Memory Library, which is software defined persistent virtual memory in the face of system crash.

Develop Inter-Process-Communication Service                                                                            Feb 2018 – Mar 2018

Building XEN style Inter-Process Communication Services Library for Domain and Clients VMs. Implementing Hypervi-

sor, Domain and Clients to test the library.

Implementation of Credit Scheduler                                                                                                   Jan 2018 – Jan 2018

Building Credit-based Thread Scheduler, which is proportional fair-share mechanism based on weight and cap.

Multi-modal Music Classification using Deep Learning                                                                 Sep 2017- Dec 2017

Main goal was to provide multi-modal classification system for music’s genre, years and gender of artist based on various

type of source(audio, lyrics and album covers). Let alone audio data which has the richest information, lyrics and album

covers are also good source of information at classifying gender of artist. Ensemble of three classification systems reinfor-

ces the final result.

Implementation and Evaluation of Deep Learning                                                                         Sep 2017 – Oct 2017

Evaluation of hyperparameters in Image Classifier using Convolutional Neural Network. Understanding the role of

hyperparameters that affects the overall performance of ML and address how to tune hyperparameters.



NISSAN RESEARCH CENTER, Sunnyvale, California                                                                  May 2018 – Aug 2018

Summer Research Intern

Connected Vehicles & Mobility Services Team


  • Implementing Data Preprocessing
  • Researching and Implementing Data Clustering
  • Implementing Ensemble Machine Learning systems for Clustered data.


“Usage Based Insurance based on Clustering and Machine Learning techniques”


LG ELECTRONICS, Seoul, Korea                                                                                                            Aug 2010 – Jul 2017

Senior Software Engineer

BSP Department, TE Group, Mobile Communications Company


  • Linux Kernel Engineer / Device Driver Developer
  • System Performance & Stability
  • System Integrating / Troubleshooting


    “V20 Android Smart Phone Project,”                                                                           Seoul, Korea (Jan 2016-July 2016)

System Integration with Qualcomm MSM8996

Troubleshooting Stability issues (L1 Cache corruption, Cache-DDR incoherency)

    “K10 Android Smart Phone Project”                                                                            Seoul, Korea (June 2015-Dec 2015)

System Integration with Qualcomm MSM8926

SDRAM/eMMC memory validation

    “Releasing Variation Project for Android”                                                               Seoul, Korea (June 2015-Dec 2015)

System Integration with Qualcomm MSM8926

    “Vu3 Android Smart Phone Project”                                                                           Seoul, Korea (Apr 2013–Dec 2013)

Troubleshooting DDR SDRAM bit flip issues.

DQS RD offset value validation (MSM8974)

    “Vu2 Android Smart Phone Project”                                                               San Diego, California (Jul 2011–Oct 2012)

Joint-Research at Qualcomm in San Diego (Aug 2012–Oct 2012)

Troubleshooting stability issues

(Bus hang, PMIC hang, sudden reset due to an access of unclocked registers)

    “B-Qwerty Android Smart Phone Project”                                                                  Seoul, Korea (Jan 2011-Aug 2011)

Implementation of device drivers for validating mass production process

(Factory Reset, DB CRC Check and AT command/DIAG protocol implementation)