• Microservices Migration - Drove pricing module of 4 services using Java Spring Boot and gRPC. Designed microservices architecture with orchestration-based saga to improve maintainability and scalability by reducing inter-service dependency. - Co-designed a Python adapter layer with an engineer to standardize 3 sets of third-party APIs. Achieved a 5X reduction in new-provider integration time, eliminating the need for case-by-case API handling from scratch.• Centralized Logging System - Designed, prototyped, and deployed a centralized logging system to collect logs from various stacks by FluentBit, Elasticsearch, and Grafana, solving issues of scattered log files and lack of dashboards. - Learned Terraform to manage GitOps for AWS infra. Contributed AWS Cognito integration to an open-source Terraform/OpenSearch project, demonstrating proactive learning and contribution beyond work. - Expanded an OpenSearch dashboard request into a comprehensive Grafana solution covering 3 data sources (CloudWatch, RDS, and OpenSearch) to meet extensive needs for dashboards and alerting mechanism.

• iOS App - Built an iOS App using React Native (TypeScript) and Redux with custom hooks for encapsulation and i18n for multi-languages, published to the App Store for commercial use, showcasing end-to-end project ownership. - Took initiative to plan 10+ product feature releases in sprints (Agile), collaborated with cross-functional teams. - Established GitOps CI/CD by GitHub Actions to automate the release process, speeding up the process by 2X. - Employed Axios interceptors to auto-refresh JWT tokens, using Queue as a throttle for concurrent requests. - Refactored pages using Factory design patterns to make pages resilient to different countries with configuration.

• Developed 2 WeChat mini-programs using Vue-like language and Rematch (Redux), handling 10+ tickets.• While handling tickets, taking initiative to migrate a module from Javascript to TypeScript with domain-driven design (DDD) structures, improved code readability and complied with the current structure guide.• Designed dynamic form to support nested structure and 3 types of input using Factory design pattern and Stack structure, decoupling business logic and codebase so form option modification does not depend on code release.

• Constructed data infra from scratch, including data warehouse ETL pipeline in Hive SQL for 40M+ records per day, customer notification module, and A/B testing framework, resulting in $370K monthly incremental revenue.• Built a user notification module with Python, boosted 5X performance through vectorization and multithreading.• Initiated flexible scripts to query and transfer data from MySQL to Hive, covering 1K+ MySQL tables.

• Created data visualization pipeline: organize data with Python script, visualized in a Tableau dashboard.• Conducted telecom fraud research, divided fraud into 4 categories, and quantified trends using Python.• Completed China consumer finance report independently, did the competitive product analysis on 3 competitive products.

• Assisted in payment module metrics by building 20+ Hive SQL templates to monitor the feature performance.• Analyzed orders' space and time distribution with a Tableau dashboard and cleaned 7M data with Python.