YuanChih(Kent) Chen
email: kent831213@gmail.com     phone: +886 919203192

| CV |
| LinkedIn |

I am a senior software engineer at Delta Electronics, entrusted with the development of an intelligent manufacturing factory automation monitoring system.

Prior to this pivotal role, I honed my skills as a software developer intern at CCRPC while pursuing a master's degree in Information Management at the University of Illinois at Urbana-Champaign.

Furthermore, I served as a C/C++ Software engineer at Systex, prior to embarking on my master's journey. I succeeded in developing Taiwan's stock trading system and provided timely and accurate stock information.

Moreover, I had several projects about distributed systems, multi-thread programming and full stack. I implemented a distributed file system. I also created a full-stack travel website using Django, JavaScript and HTML5, and wrote store procedures and triggers to make the applications comprehensive.

Based on the above experience, I am eager to pursue a software engineering career. I’m looking for an opportunity where I can contribute with my diverse experiences and abilities.

sym


DELTA
Senior Software Engineer
Dec. 23 - Now

sym


CCRPC
Software Developer Intern
Jul. 22 - May 23

sym

UIUC
MSIM
Aug. 21 - May. 23

sym

Systex
Software Engineer
Dec. 20 - Apr. 21

sym

TKU
B.B.A in IM
Sep. 16 - Jun. 19

  News
  Publications
sym

Artificial Intelligence for Conversational Robo-Advisor
Min-Yuh Day,Jian-Ting Lin, YuanChih Chen
IEEE/ACM (Oral Presentation)

| abstract | pdf |

With the upcoming of the AI era, AI combines withFintech (Financial Technology; FinTech) has become a development tendency in the financial industry. However,the literatures are not as common investigated into deep learning about the application of automated financialmanagement. This research focuses on the application of Fintech and deep learning in asset allocation, and expectsto optimize the investment portfolio. The best investing in portfolio Index-based funds based on Taiwan's index-typesecurities investment trust funds (ETFs) are the main investment targets, and the time series related models fordeep learning – Long Short Term Memory predicts the increase of each investment target, and finds the bestinvestment portfolio in combination with the relevant asset allocation theory; the HCI dialogue service adopts AIMLand a Generative model. The AIML achieves the effect of dialogue by making a knowledge, and the generated modelis completed by the Sequence To Sequence model. In this research, we use the Markowitz Mean-Variance Model andBlack-Litterment as our allocation theories, the result showing that the Blacklitterment model had a betteraccumulated return performance and outperformed other strategies.

            @article{chao18radiotherapy,
                title     = {Radiotherapy Target Contouring with Convolutional Gated Graph Neural
                             Network},
                author    = {Chao, Chun-Hung and Cheng, Yen-Chi and Cheng, Hsien-Tzu and Huang, Chi-Wen and
                             Ho, Tsung-Ying and Tseng, Chen-Kan
                            Lu, Le and Sun, Min},
                journal   = {arXiv preprint arXiv:1904.02912},
                year      = {2019},
                }
            
  Projects

Mulit Containers Project
• Developed a web application utilizing Nginx, React, Express.js, AWS Elastic Beanstalk, S3, and RDS.
• Built multiple containers with Nginx, React, Express, and Redis by using Docker and Docker Compose.
• Implemented an automated CI/CD pipeline using Travis CI for build, testing, and deployment to AWS Elastic Beanstalk.

Distributed Machine Learning System
Created a multi-client machine learning system supporting fair-resource training, fair-time inference, and fault tolerance by using a Leader-based Consensus protocol. And designed an algorithm to make the query processing rates within 20% between AlexNet and ResNet18.

Distributed File System
Implemented a Paxos-like protocol for replicated files containing Leader Election, file replication, and file persistence to improve fault tolerance in the cluster.

Distributed Group Membership
Implemented a Gossip-style protocol for members can failure detect and disseminate the newest group status to all other members to support fault tolerance, completeness, and scalability in the cluster.

Log Querier System
Created a multi-client log querier system supporting fault-tolerant to ensure Consistency and Isolation of distributed log querier by using socket connections and C++.

Communication Network
• Implemented an HTTP client-server connection and transfer data from server to client correctly.
• Designed a TCP protocol algorithm to achieve flow control, congestion control, and achieve reliable data transfer by using UDP connection between client and server.
• Implemented link state and distance vector routing protocols to finish routing tables, including tie-breaking and minimum cost path.

Travel Assistant
Utilized Python, Django, and frontend languages to design a travel assistant website.
| website demo video |

Using Deep Learning in Conventional-Robo-Advisor
A website provided the timely weather, stock's information, and Customized investment portfolio recommendations.
| website demo video | pdf |

Line chatbot
Used Twitter and movie conversations as corpus to implement natural language processing.

  Honor and Awards
  • 1st Place, International ICT Innovative Services Competition, Chinese Society for IM, Taiwan
  • 1st Place, Department of Information Management Project Competition, TKU
  • 2nd Place, MakeNTU Hackathon Competition, NTU
  • 3rd Place, JihSun Bank Hackathon Competition
  • Undergraduate Academic Excellence Award, TKU

Template: this