Linux platforms, applying them to manual and automated testing solutions for product functionality and security. Taoyuan City, Taiwan Certifications and related courses Information Security Certification ISC2 CC (certified in cybersecurity) ISC2 CISSP (Certified in Information Security Professional) EC-Council CEH IT System-related and other certifications RedHat RHCSA RedHat RHCE Digital Electronic Circuit Inspection Class B Certification. Security Certifications and Training Courses ISO 27001:2022 Lead Auditor TUV ISA/ IECtraining TWCSA - RAT Advanced Red Team Training. EducationConcordia University Master of Business Administration (M.B.AMinghsin University of Science and Technology (Two-Year College) Electronic
國立中央大學 National Central University 資訊工程學系 資格認證 金色證書 多益TOEIC 發照日期 九月 2017 · 永久有效 CCNA Cisco 發照日期 四月 2010 · 永久有效 RHCE Red Hat 發照日期 六月 2008 · 永久有效 LPIC-1 Linux Professional Institute 發照日期 三月 2007 · 永久有效 專案 宅神爺 宅神爺為一款合法的線上紙牌與麻
.跑合作廠商Microsoft PC & Printer 定期保養 3.智慧局OP每月固定代班及有OP請假代班, 智慧局工作性質是專利資料轉換及入資料庫 學歷龍華科技大學LUNGHWA UNIVERSITY OF SCIENCE AND TECHNOLOGY 電機工程系 資格認證 Red Hat Certified Specialist in Virtualization redhat九月 2022 到期 RHCE redhat九月 2022 到期 RHCSA redhat九月 2022 到期
Word
Excel
Microsoft Office
Employed
・
Open to opportunities
Full-time / Interested in working remotely
4-6 years
龍華科技大學LUNGHWA UNIVERSITY OF SCIENCE AND TECHNOLOGY
Monitoring and alerting services with Prometheus, AlertManager and Grafana * Able to clarify situations when system-level errors occur (Ops, Dev) Education Master of Computer Science and Information Engineering National Changhua University of EducationBachelor of Electronic Engineering, System Applications Group Southern Taiwan University of Science and TechnologyCertification RED HAT CERTIFIED ENGINEER (RHCE) Red Hat Credential ID#linux> Credential Link Skill Project A Smart meter Design implemented with IOT The Node.js-based platform collects and monitors data from clients with LinkIt Smart 7688 to detect electrical power events. * Capture power-signal with Golang and Linklt Smart 7688 Duo
JavaScript
Node.js
Kubernetes
Employed
Full-time / Interested in working remotely
4-6 years
National Changhua University of Education
・
Master of Computer Science and Information Engineering
網路架構規劃、客戶服務、故障排除、硬體諮詢 Education恆逸資訊 雲端虛擬化系統工程師就業認證養成班大葉大學 工業工程與科技管理 IT 相關證照 CCNA Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. RHCE Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh. MCSA Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh.
National Chenchi University, MS, Statistics, 2015 – 2017
GPA : 3.84 / 4.0
Master Thesis: Entropy Based Feature Selection, Professor Pei-Ting, Chou
Objective: Build a similarity matrix based on Mutual Entropy under Hierarchical Clustering. Afterwards, select clustered features as the final selection.
Compare the model with other feature selection methods like RF, Lasso, F-score.
National Chen-Kung University, BS, Mathematics, 2011 – 2015
Skills
Programing
Python
Scala
R
MSSQL
Data-related Tools
Tensorflow (Keras)
PyTorch
Spark
Docker
Scikit-Learn
Pandas
Cloud Platform
AWS
GCP
Language
English: TOEFL 98 / 120
Work Experience
CTBC Bank, Model Development Department, Data Scientist
2021.12 – present
About the department:
Responsible for developing models related to bank recommendations and risks, including projects such as coupon recommendations, account opening marketing lists, and fraud detection.
Job responsibilities:
Throughout the entire project lifecycle, my primary responsibilities included model design, model training, end-to-end process development, feature design, performance tracking, and method research.
Fraud Alert Project
Objective:
Predicting potential fraudulent accounts based on transaction data, restricting transactions in advance to prevent harm.
Responsibilities/Achievements:
Development and deployment of credit card and financial features.
Managing the data flow process from receiving variables to model predictions, identifying risk factors, and updating alert lists.
Implemented Autoencoder + contrastive learning to achieve a 1.81% improvement in model effectiveness.
Coupon Recommendation
Objective:
Personalized coupon recommendations for mobile banking users to increase click-through rates and redemption rates.
Responsibilities/Achievements:
Utilized multi-task learning to simultaneously predict click-through behavior and coupon redemptions, resulting in a 14% increase in click-through rate and a 74% increase in redemption rate.
Created performance tracking reports to monitor online model performance and provide insights to Business Units.
Financial Product Recommendations
Objective:
Tailored financial product recommendations for mobile banking users to enhance click-through rates without compromising conversion rates.
Responsibilities/Achievements:
Applied multi-task learning to jointly learn click-through and conversion behaviors, fine-tuned model architecture, achieving a 90% outperformance against competitor models in online testing.
Marketing List for Digital Savings Accounts
Objective:
Optimized conversion rates for marketing lists related to digital savings accounts
Responsibilities/Achievements:
successfully raising conversion rates from 0.23% to 1.16%
Work Experience
CLICKFORCE, Data Engineer Supervisor, 2020.1 – 2021.11
About the company:
As a top domestic digital advertisement company, CLICKFORCE cooperates with over 900 web media and over 400 mobile media to build a huge advertising environment. CLICKFORCE considers data-driven solution as the core concept of the company, and dedicates to help advertisers to achieve their commercial goals.
At 2020, CLICKFORCE won 2 awards at Agency & Advertiser of the Year.
Successfully acquire the exclusive advertising agency qualification for Tokyo 2020 Olympics in Taiwan.
Job responsibilities:
Optimize ad performance from all aspects, including the system, target audience tags, etc.
Do researches for new ML model (recommender model, NLP model) or architecture which is suitable for our system.
Develop data-related products or projects.
Analyze data to help improve our system or inspect whether the demands from business side is doable.
Real-time AD Recommender System
Objective:
Building a real-time ad recommender system to upgrade our ad server and get better performance.
Responsibilities:
Figure out what kind of recommender system components that is suitable for our ad system.
Build a tower-like and feature-cross model refer to other famous recommender system model.
Responsible for system engineering, which includes data preprocessing, embedding generates, memory cache, cold start, model API, etc.
Interest Tags
Objective:
Build interest tags for ads to help ad optimizers choose their target audience.
Responsibilities:
Create the features from what articles they saw, what website they viewed, and what ads they interacted.
Deal with 20 million rows data and 120 million inference samples.
Build ML model to predict each user's behavior on certain ads.
Using Spark through AWS EMR to accelerate the speed of producing tags.
Achievements:
Raise CTR performance up to 200-300% of the original tags depends on different tags, and gain more impression while maintain better performance.
After accomplishing this project, we terminated the cost on purchasing interest tags from other company, and successfully turned the original cost into revenue by providing profitable data.
First Party Cookie Mapping
Objective:
Deal with the Google 3rd party Cookie issue, figure out a method to map numerous 1st party Cookies to a user.
Responsibility:
Transform this problem into a ML mission. Design the label of the data, figure out what feature we can get or produce and whether the feature is useful for the goal.
Apply XGboost on this mission.
Build a small test to prove this method works.
Achievement:
70% of precision.
One of the solution of our company while the cancelation of 3rd party Cookie happen.
Invoice Data Application
Objective:
Develop invoice data application.
Responsibility:
Responsible for fine-tuning BERT to predict category for each product.
Produce invoice data report to brands or business unit. It demonstrates the sales volume across different channel, what kind of products are frequently bought together, and also shows comparison of target brand to the other brands.
Achievements:
Produce an invoice data report product.
Produce invoice tags for ad system.
Other Experience
E.Sun AI 2020 Summer Competition, 2020.7 – 2020.8
Objective:
Extract names of money laundering suspects from an article.
Responsibilities:
Crawl the articles from different media, and parse them by using Selenium, Requests, and Beautiful Soup.
Construct 2-step model: First, identify whether the article is related to money laundering. Second, extract the suspects' names.
Build model serving API by Tensorflow Serving.
Build REST API for preprocessing request data and return the prediction.
Achievement:
23rd place among 409 teams.
Youtube Data-Driven Marketing System, Institute for Information Industry, 2019.8 – 2019.11
Objectives:
Use the title and the description of videos to automatically classify videos.
Use the title and the description of videos to identify whether a video is sponsored.
Give suggestions for Youtubers or companies who desire to sponsor in a video based on data analysis.
Responsibilities:
Apply Google API and write Python functions to get structured raw data.
Train word vectors using Gensim based on Wiki's open data.
Use the frequency of each sentence as a criteria to eliminate useless words.
Tune LSTM, Conv1D, BERT on the NLP mission.
Use EDA methods to see the insights of the data under different classes and different sponsored status.
Achievement:
71% accuracy in classifying video’s type.
89% accuracy in detecting sponsored content.
E.Sun Real Estate Price Prediction Competition, 2019.7 – 2019.8
Objective:
Use the real estate training data to build a model and predict the real estate price within 10% residual.
Responsibilities:
Apply XGBoost, LGBM and other ML models to train the model.
Collect the outputs as new features from each ML model and add them into the original data set to enhance the performance of the final model.
Achievement:
150th place out of 1200 teams.
KKTV Data Game,2017.5 – 2017.6
Objective:
Predict the next video a user watch in the next time interval.
Responsibilities:
Extract different features from raw data, such as the latest video, the video which got the longest viewing time, the video which got the largest number of viewing.
Use the user viewing data to construct a similarity matrix of each video as additional features.
Achievement:
10th place out of 50 teams.
MRT Open Data Competition, 2017.4 – 2017.5
Objective:
Study the changes of passenger volume of MRT by surrounding geometric data.
Responsibilities:
Apply bisection method to build the edges between MRT stations.
Combine other geometric data based on these borders.
Use Lasso feature selection method to explore the importance of each feature.
Add noises into features to check the features are not randomly selected.
National Chenchi University, MS, Statistics, 2015 – 2017
GPA : 3.84 / 4.0
Master Thesis: Entropy Based Feature Selection, Professor Pei-Ting, Chou
Objective: Build a similarity matrix based on Mutual Entropy under Hierarchical Clustering. Afterwards, select clustered features as the final selection.
Compare the model with other feature selection methods like RF, Lasso, F-score.
National Chen-Kung University, BS, Mathematics, 2011 – 2015
Skills
Programing
Python
Scala
R
MSSQL
Data-related Tools
Tensorflow (Keras)
PyTorch
Spark
Docker
Scikit-Learn
Pandas
Cloud Platform
AWS
GCP
Language
English: TOEFL 98 / 120
Work Experience
CTBC Bank, Model Development Department, Data Scientist
2021.12 – present
About the department:
Responsible for developing models related to bank recommendations and risks, including projects such as coupon recommendations, account opening marketing lists, and fraud detection.
Job responsibilities:
Throughout the entire project lifecycle, my primary responsibilities included model design, model training, end-to-end process development, feature design, performance tracking, and method research.
Fraud Alert Project
Objective:
Predicting potential fraudulent accounts based on transaction data, restricting transactions in advance to prevent harm.
Responsibilities/Achievements:
Development and deployment of credit card and financial features.
Managing the data flow process from receiving variables to model predictions, identifying risk factors, and updating alert lists.
Implemented Autoencoder + contrastive learning to achieve a 1.81% improvement in model effectiveness.
Coupon Recommendation
Objective:
Personalized coupon recommendations for mobile banking users to increase click-through rates and redemption rates.
Responsibilities/Achievements:
Utilized multi-task learning to simultaneously predict click-through behavior and coupon redemptions, resulting in a 14% increase in click-through rate and a 74% increase in redemption rate.
Created performance tracking reports to monitor online model performance and provide insights to Business Units.
Financial Product Recommendations
Objective:
Tailored financial product recommendations for mobile banking users to enhance click-through rates without compromising conversion rates.
Responsibilities/Achievements:
Applied multi-task learning to jointly learn click-through and conversion behaviors, fine-tuned model architecture, achieving a 90% outperformance against competitor models in online testing.
Marketing List for Digital Savings Accounts
Objective:
Optimized conversion rates for marketing lists related to digital savings accounts
Responsibilities/Achievements:
successfully raising conversion rates from 0.23% to 1.16%
Work Experience
CLICKFORCE, Data Engineer Supervisor, 2020.1 – 2021.11
About the company:
As a top domestic digital advertisement company, CLICKFORCE cooperates with over 900 web media and over 400 mobile media to build a huge advertising environment. CLICKFORCE considers data-driven solution as the core concept of the company, and dedicates to help advertisers to achieve their commercial goals.
At 2020, CLICKFORCE won 2 awards at Agency & Advertiser of the Year.
Successfully acquire the exclusive advertising agency qualification for Tokyo 2020 Olympics in Taiwan.
Job responsibilities:
Optimize ad performance from all aspects, including the system, target audience tags, etc.
Do researches for new ML model (recommender model, NLP model) or architecture which is suitable for our system.
Develop data-related products or projects.
Analyze data to help improve our system or inspect whether the demands from business side is doable.
Real-time AD Recommender System
Objective:
Building a real-time ad recommender system to upgrade our ad server and get better performance.
Responsibilities:
Figure out what kind of recommender system components that is suitable for our ad system.
Build a tower-like and feature-cross model refer to other famous recommender system model.
Responsible for system engineering, which includes data preprocessing, embedding generates, memory cache, cold start, model API, etc.
Interest Tags
Objective:
Build interest tags for ads to help ad optimizers choose their target audience.
Responsibilities:
Create the features from what articles they saw, what website they viewed, and what ads they interacted.
Deal with 20 million rows data and 120 million inference samples.
Build ML model to predict each user's behavior on certain ads.
Using Spark through AWS EMR to accelerate the speed of producing tags.
Achievements:
Raise CTR performance up to 200-300% of the original tags depends on different tags, and gain more impression while maintain better performance.
After accomplishing this project, we terminated the cost on purchasing interest tags from other company, and successfully turned the original cost into revenue by providing profitable data.
First Party Cookie Mapping
Objective:
Deal with the Google 3rd party Cookie issue, figure out a method to map numerous 1st party Cookies to a user.
Responsibility:
Transform this problem into a ML mission. Design the label of the data, figure out what feature we can get or produce and whether the feature is useful for the goal.
Apply XGboost on this mission.
Build a small test to prove this method works.
Achievement:
70% of precision.
One of the solution of our company while the cancelation of 3rd party Cookie happen.
Invoice Data Application
Objective:
Develop invoice data application.
Responsibility:
Responsible for fine-tuning BERT to predict category for each product.
Produce invoice data report to brands or business unit. It demonstrates the sales volume across different channel, what kind of products are frequently bought together, and also shows comparison of target brand to the other brands.
Achievements:
Produce an invoice data report product.
Produce invoice tags for ad system.
Other Experience
E.Sun AI 2020 Summer Competition, 2020.7 – 2020.8
Objective:
Extract names of money laundering suspects from an article.
Responsibilities:
Crawl the articles from different media, and parse them by using Selenium, Requests, and Beautiful Soup.
Construct 2-step model: First, identify whether the article is related to money laundering. Second, extract the suspects' names.
Build model serving API by Tensorflow Serving.
Build REST API for preprocessing request data and return the prediction.
Achievement:
23rd place among 409 teams.
Youtube Data-Driven Marketing System, Institute for Information Industry, 2019.8 – 2019.11
Objectives:
Use the title and the description of videos to automatically classify videos.
Use the title and the description of videos to identify whether a video is sponsored.
Give suggestions for Youtubers or companies who desire to sponsor in a video based on data analysis.
Responsibilities:
Apply Google API and write Python functions to get structured raw data.
Train word vectors using Gensim based on Wiki's open data.
Use the frequency of each sentence as a criteria to eliminate useless words.
Tune LSTM, Conv1D, BERT on the NLP mission.
Use EDA methods to see the insights of the data under different classes and different sponsored status.
Achievement:
71% accuracy in classifying video’s type.
89% accuracy in detecting sponsored content.
E.Sun Real Estate Price Prediction Competition, 2019.7 – 2019.8
Objective:
Use the real estate training data to build a model and predict the real estate price within 10% residual.
Responsibilities:
Apply XGBoost, LGBM and other ML models to train the model.
Collect the outputs as new features from each ML model and add them into the original data set to enhance the performance of the final model.
Achievement:
150th place out of 1200 teams.
KKTV Data Game,2017.5 – 2017.6
Objective:
Predict the next video a user watch in the next time interval.
Responsibilities:
Extract different features from raw data, such as the latest video, the video which got the longest viewing time, the video which got the largest number of viewing.
Use the user viewing data to construct a similarity matrix of each video as additional features.
Achievement:
10th place out of 50 teams.
MRT Open Data Competition, 2017.4 – 2017.5
Objective:
Study the changes of passenger volume of MRT by surrounding geometric data.
Responsibilities:
Apply bisection method to build the edges between MRT stations.
Combine other geometric data based on these borders.
Use Lasso feature selection method to explore the importance of each feature.
Add noises into features to check the features are not randomly selected.