- Performed cloud migration of quant factor computation pipeline and data from an in-house SQL Server to AWS S3 + Parquet to improve service availability and to cut down costs. Developed ETL scripts using Python and partitioned Parquet data based on SQL table structure to improve data access efficiency by 20%.- Solved the consistency issue of concurrent reading and writing factor data in the development environment. Enhanced the task executor by adding distributed locks to the data and… Show more - Performed cloud migration of quant factor computation pipeline and data from an in-house SQL Server to AWS S3 + Parquet to improve service availability and to cut down costs. Developed ETL scripts using Python and partitioned Parquet data based on SQL table structure to improve data access efficiency by 20%.- Solved the consistency issue of concurrent reading and writing factor data in the development environment. Enhanced the task executor by adding distributed locks to the data and added logging capabilities to ensure data consistency and task observability. Show less

- Developed three data visualization platforms for risk control and compliance department using React. Implemented features like role-based access control and custom dashboards. Optimized the performance of the web page with lazy loading, code splitting, pagination, memo, gzip compression, etc. Reduced First Contentful Paint time from 2s to 280ms, compressed bundle size by 78%. - Implemented a low-code chart library based on ECharts to simplify manual configuration and improve component… Show more - Developed three data visualization platforms for risk control and compliance department using React. Implemented features like role-based access control and custom dashboards. Optimized the performance of the web page with lazy loading, code splitting, pagination, memo, gzip compression, etc. Reduced First Contentful Paint time from 2s to 280ms, compressed bundle size by 78%. - Implemented a low-code chart library based on ECharts to simplify manual configuration and improve component reusability. Encapsulated data query platform for 8 commonly used charts (bar, line, pie, etc.), which supports querying and rendering data by id or custom logic in front-end. Reduced configuration LOC by 83% and has been applied to more than 200 metric charts.- Enabled automatic chart export to editable PowerPoint feature using pptxgenjs. Uniformed styles across web and slides to improve user experience. Solved the pain point of manual work under massive chart usages and has been applied to more than 200 charts. Show less

Research assistant for Professor Lihua Xu, in mutation testing in the area of financial technology.- Investigated the application of mutation testing technology in the field of financial technology.- Created a customized mutation operator under the PiTest framework with Java ASM, which supports specific code pattern recognition and mutation on it. - Collaborated with teammates, and conducted test case validity experiments provided by an automated test generation suite.

Learning Assistant for Introduction to Computer Programming, cooperating with Professor Romain Corcolle.

- Developed an automatic hedging cost calculation and storage tool using Python and MySQL.- Implemented a web crawler with proxies for EastMoney stock community forum (a professional Chinese financial information platform) using Python-based spiders and MySQL for storage.