工作經歷
樂居科技, 數據組組長, 2022/7~迄今
- Airflow引進：開發新的DAG，並將舊有系統都導入到Airflow，實現事件驅動的數據處理流程。
- 雲端服務優化：運用AWS和GCP的雲端服務，如SQS、Parameter Store、Cloud Function和Lambda，提高現有系統的穩定性。
- API開發：與前端、後端規劃API規格，使用FastAPI、DuckDB和Docker完成樂居網站專業版的後端API，使此API能在安全且高效的情況下，將數據傳送到前端呈現給用戶。
- 效能優化：針對系統瓶頸進行優化，縮短程式約30%的執行時間，提升數據處理效率。
- 團隊領導：帶領兩位數據工程師和三位實習生，與PM及前後端團隊合作，負責系統架構設計、需求解決、時程估算，並維持團隊開發進度，以達成專案目標。
樂居科技, 數據工程師, 2020/7~2022/6
- 資料管線建立：建立了公司的資料管線，整合不同資料來源的數據至Bigquery，解決同仁資料取得不易的問題。
- 自動化系統：建立自動化資料系統，節省公司人力99%的重複性作業時間，提升效率。
- 監控儀表板：使用Looker Studio，建立系統監控的儀表板，提高系統的透明度，使團隊能更精準掌握系統運行狀況。
- 文化導入：導入Code review、TDD，提高程式維護性及可靠度，進而提升團隊的產能。
- 部屬優化：利用CI/CD，縮短了每次進版的時間，減少人工時間和人為錯誤。
- 雲端服務運用：善用Cloud Service，如EC2、RDS和Compute Engine，降低維運成本。
研華科技 Advantech, 研發工程部實習生, 2015/7 ~ 2015/8
- Alljoyn 框架應用：利用 Alljoyn 框架進行概念驗證 (POC)，將公司的 WISE-PaaS 設備資訊從手動輸入轉為自動輸入。這項改變不僅減少了錯誤率，還節省了人力成本和時間。
學歷
台灣大學, 碩士學位, 資訊工程, 2015/9 ~ 2018/8
於Newslab接受施吉昇教授指導，實驗室領域包含：作業系統研究、排程理論研究、無線網路、嵌入式系統、異質計算系統、雲端系統，與行動運算中介軟體。
- 深耕計畫 | 2015/9 - 2017/9
- 計畫名稱：異質無線端點時序資料同步技術
- 目標： 集結台灣大學、交通大學、台灣科技大學、中正大學等各種領域的團隊， 利用工業4.0的概念，開發出自動化平台。
- 成果：伺服器收集到的各種感測器的感測資料可以依據感測的時間順序排列
- 論文 "在控制器區域網路匯流排之裝置選擇機制及資料時序校正"
台北市立大學, 學士學位, 資訊科學, 2011/9 ~ 2015/6
- Kaggle Competition：CIFAR-10 – Object Recognition in Images: Identify the subject of 300,000 labeled images
- 比賽目標：對30萬張的32x32的彩色圖片進行分類，共有10種類別。
- 比賽成果：使用深度學習進行分析，得到全球前30%的名次。
- 以此比賽作為畢業專題，於103年度的畢業專題展覽得到特優。
- 103年度社群網路與應用課程之專題第一名
技能
專業技能
- Python, SQL
- GitHub Flow
- Google Cloud Services
- AWS Cloud Service
- Data Warehouse
- Data Pipeline
- Test-Driven Development
- Scrum
- Docker
- CI/CD
- Airflow
英文熟練度
TOEIC - Reading 375
- Listening 320
Bo-Wen Xiao
Cell phone : 0956154872
Mail：[email protected]
Blog： https://medium.com/i-think-so-i-live
Position：Software Engineer
Work Experience
Leju Technology, Data Team Lead, 2022/7~Present
- Import Airflow: Developed new DAGs and migrated existing systems to Airflow, enabling event-driven data processing workflows.
- Cloud Service Optimization: Leveraged AWS and GCP cloud services, including SQS, Parameter Store, Cloud Function, and Lambda, to enhance the stability of existing systems.
- API Development: Collaborated with front-end and back-end teams to implement the back-end API for Leju’s professional website using FastAPI, DuckDB, and Docker. This API securely and efficiently delivers data to the front-end for user presentation.
- Performance Optimization: Utilized performance profiling tools to optimize system bottlenecks, resulting in reduced program execution time and improved data processing efficiency.
- Team Leadership: Led a team of two data engineers and three interns. Worked closely with PMs and cross-functional teams to design system architecture, address requirements, estimate project timelines, and maintain development progress to achieve project goals.
LejuTech, Data Engineer, 2020/7~2022/7
- Data Pipeline Establishment: Created the company’s data pipeline, integrating data from various sources into BigQuery, resolving data accessibility challenges for colleagues.
- Automation Systems: Developed automated data systems, significantly reducing repetitive manual tasks by 99% and enhancing overall efficiency.
- Monitoring Dashboard: Leveraged Looker Studio to build a system monitoring dashboard, improving transparency and enabling the team to better understand system performance.
- Cultural Adoption: Introduced code reviews and TDD practices to the team, enhancing code maintainability, reliability, and overall team productivity.
- Deployment Optimization: Implemented CI/CD processes, reducing deployment time and minimizing human errors.
- Cloud Service Utilization: Effectively utilized cloud services such as EC2, RDS, and Compute Engine, resulting in reduced operational costs
Advantech, Summer intern, 2015/7 ~ 2015/8
- Application of Alljoyn Framework: Conducted a Proof of Concept (POC) using the Alljoyn framework to automate the input of WISE-PaaS device information, replacing manual data entry. This change not only reduced error rates but also saved on labor costs and time.
Education
National Taiwan University, Master Degree, Department of Computer Science & Information Engineering, 2015/9 ~ 2018/8
I joined the Newslab under Professor Chi-Sheng Shih's. The research area of this lab contains operation system, scheduling theory, wireless network,embedded system, cyber-physical systems and cloud system.
- PaaS project | 2015/9 - 2017/9
- Subproject name : Data Time Synchronization from Cyber-Physical Wireless Node.
- Goal : This project members included professors and students in different research areas from NTU, NCKU, NCTU, CCU and Taiwan Tech. This project aimed to develop a platform for Industry 4.0.
- Results : Sensed data received by server is arranged in the order in which they sampled.
- Thesis "Device Selection on CAN bus for Temporal Data Alignment"
University of Taipei, Bachelor Degree, Department of Computer Science ,
2011/9 ~ 2015/6
- Kaggle Competition：CIFAR-10 – Object Recognition in Images: Identify the subject of 300,000 labeled images
- Goal : To predict CIFAR-10 dataset which consists of 300,000 32*32 color images with one of 10 object classes.
- Results : I used deep learning to analysis and ranked in the top 30% in the world.
- I chose image recognition as my graduation project and got the Distinguished Honor Award in the graduation project exhibition.
- First place of Social Network & Application project in 2014.
Skills
Professional Skills
- Python, SQL
- GitHub Flow
- Google Cloud Services
- AWS Cloud Service
- Data Warehouse
- Data Pipeline
- Test-Driven Development
- Scrum
- Docker
- CI/CD
- Airflow
English Proficiency
TOEIC - Reading 375
- Listening 320
