Certification
- AWS Certified Developer – Associate
This badge was issued to Yu Chen Hsieh on February 27, 2023 Expires on February 27, 2026.
Work Experience
2022/09 - 2023/09 Cloud Engineer, Kiwi technology Inc.
- Refactor existing monolithic service to cloud-native architecture
- Solve several n+1 query problems
- Build CI/CD pipeline
- Build Log System with Loki+Grafana
2020/04 - 2022/05 Backend Engineer,Groundhog Technologies
- Develop Graphql API in nodejs
- Experience in benchmark measurement
- Experience using Ansible to deploy services
- Experience in full stack development
- Implement authorization service
- Use performance profiling to find bottleneck of our service. Reduce response latency 22%
- Integrated with native ads
- Use Protobuf to reduce data transmission across the network
2019/09 - 2020/03 Backend Engineer, Roadmio Crop.
- Design the new system architecture with coworkers
- Integrate 3rd party APIs, includes taiwanmobile and every8D (text message)
- Import AWS ECS to new system
- Implement the deployment tool using AWS cdk
- Implement 3 microservices
- Setup ssh tunnel from local server to jump server
Skills
Language
- Node JS
- Python
Database
- MongoDB
- Redis
- SQLite
Operator
- Git CI/CD
- Ansible
- Docker
- Grafana
- Loki
Cloud
- AWS
(ECS,ECR,SQS,cdk, etc.)
Other
- Kafka
- MQTT
- Protobuf
- Jest
Education
2017/09 - 2019/07 National Tsing Hua University, Taiwan (Suspension)
- M.S., Department of Computer Science
- Thesis: Joint Virtual Image Download and Rate Allocation of Internet-of-Things Analytics
- Abstract: Smart city is probably the most popular usage scenario of IoT, in which IoT analytics allows better use of resources, higher Quality-of-Service, and lower management cost. The present solution would run all the IoT analytics on the data center. But this may cause: (i) excessive Internet access cost and (ii) degraded QoS. To solve this problem, we suggest running some of the IoT analytics on gateways. Moreover, we proposed 2 algorithms to determine (i) which IoT analytics should run on the gateway and (ii) each IoT analytics should consume how much bandwidth.
2013/09 - 2017/06 National Central University, Taiwan
- B.S., Department of Atmospheric Science
Publication
- Y. Hsieh, H. Hong, P. Tsai, Y. Wang, Q. Zhu, M. Uddin, N. Venkatasubramanian, and C. Hsu, “Managed Edge Computing on Internet-of-Things Devices for Smart City Applications”, ” in Proc. NOMS IEEE/IFIP Netw.
Oper. Manag. Symp., Apr. 2018, pp. 1–2.
School Project ：Managed Edge Computing on Internet-of-Things Devices for Smart City Applications
We build a managed edge computing platform for Internet-of-Things (IoT). This platform consists of a model city, multiple Raspberry Pis as minions, and a mini PC as a master. Through the web dashboard served on the master, we can remotely monitor, manage, and upgrade the IoT devices and IoT analytics running on the devices. Once a user sends a request to our system, our system will deploy the related application to the device which is near to users. In this project, I have learned the following open sources:
- Kubernetes: devices resource management, docker container deployment
- Docker: multi-tenancy, resource reservation
- TensorFlow: distributed analytics
- MQTT: transmit sensor data
https://www.youtube.com/watch?v=RFe_D6Mkdhw
School Project ：New Music Play Mode for Karaoke
The new music play mode can make the karaoke song list more flexible and the users get a better singing experience. Usually, there are many users in a karaoke box, but there are only one or two singers. Our music play mode plays the next music according to recognizing the present singer instead of the order of the song list. We implement this system with AWS, web page, and Raspberry Pi. In this project, I am responsible for the following functions:
- Cognito authority management
- Create DynamoDB table from front-end
- Publish/subscribe messages to Raspberry Pi from back-end using IoT core
- Pass/receive messages using SQS/SNS from front-end/back-end
- Write a web page to dynamic display Youtube videos
