- Utilized DBT for Data Lineage to reduce the complexity of our ETL pipeline which improved the development process by 30% - Simplified the ETL pipeline and execution dependencies using Docker - Migrated tasks that require heavy resources to Snowflake, resulting in reduced execution time for scheduled tasks by 90% and enhanced convenience for data scientists - Survey Snowflake and compare it to our system at that time and make an internal sharing to other development teams later - Built training and monitoring process for the price forecasting model on the AWS Sagemaker - Built internal reports for the finance, operation, and product team to know better our own products, and also built external report dashboards for our customers' daily usage

Data Engineering and Machine Learning - Constructed a streaming data pipeline which is capable of handling 600k events/min in a 2-core CPU environment for an anomaly detection system - Developed data preprocessing tools using PySpark on AWS to cater various data scientist requirements - Created time-series anomaly detection systems and log-based detection systems Development - Packaged and deployed data pipeline and model serving services to Kubernetes - Efficient component of streaming data pipeline with Golang

Data Engineering - Developed, maintained, and optimized a data pipeline which increases service availability from 95% to 99.9% while handling 10 million events per day - Activity analyzed and visualized data to enhance decision-making for team leaders and product managers - Built performance and monitoring dashboards and utilize standard deviation to showcase the trends in our metrics which helps us to understand the performance of our product - Developed and maintained the ETL pipeline with Airflow and PySpark Machine Learning - Constructed a text CNN model using PyTorch for precise advertising article classification - Developed models for generating article embeddings to enhance the recommender system's capability in suggesting related articles Development - Implement data analysis component with backend engineers - Deployed programs using Docker to enhance maintainability and flexibility

Development - Developed backend applications and designed database schema - Integrated third-party payment solutions into the payment procedures, resulting in a 10% increase in revenue - Implemented the Git workflow and coding style in the development process to decrease the occurrence of code conflicts and human errors by 60% Data Engineering and Machine Learning - Transplant search system with Elasticsearch that reduces cost by 75% - Data analysis and visualization for testing the feasibility of new product - Build machine learning for spam article filter

- Built websites with PHP and JavaScript