• Accomplished an 80% faster deployment efficiency by designing a region-based deployment infrastructure
pipeline for building test packages in new commercial/isolated regions for our web service.
• Deployed test packages to all 34 AWS global regions (108 availability zones) for our web service, resulting
in 100% global service coverage.
• Minimized deployment failures to 3% with wave deployment and auto version rollback, resulting in 97%
uptime.
• Achieved a 99% API test coverage by implementing continuous testing to mimic customer interactions with
service endpoints, resulting in a 10% higher SLA regarding reliability.
• Reduced 60% by factoring infrastructure code, allowing for IPv6 enablement and future extensibility.

• In charge of adding new features to the admin panel used by moderators and administrators
• Revamped React.js code from the ground up for better extensibility and adhered to the SOLID principles, reducing code-based by 33%
• Created automated unit tests and integration tests for detecting user-related issues, improving bug detection by 10%

• Enhanced system efficiency by 30% by developing new API features and removing unnecessary API calls,
whilst strictly adhering to SOLID principles and clean coding standards.
• Increase API test code coverage by 30% to cover new feature interactions and edge cases, resulting in a more
robust system and service.
• Increase testable code by 50% by creating injectable system variables classes to mock each service class.

• Led 4 developers following Agile principles to build a portfolio management system, allowing customers to
view financial data, news, and more, improving customer satisfaction by 40% through client interaction.
• Contributed up to 8,000 lines of code to implemented domain models/API endpoints for authentication,
profile management, portfolio tracking, and AI chatbot functionalities, using FastAPI and Peewee ORM.
• Achieved 85% code coverage to ensure the robustness of APIs and service methods through performing rigorous unit tests, API call mocking, and integration tests.

• Collaborated with cross-functional teams to create predictive models using Python libraries such as NumPy and Pandas, reducing overdue amounts by $70M.
• Performed statistical analyses such as principle component analysis(PCA) to identify areas for improvement that resulted in a 15% reduction in operational costs.
• Automated report generation using Matplotlib, Scipy, Numpy, Pandas, and python-docx while adhering to object-oriented design principles and increased efficiency by 10 times.

• Cooperated with 8 others to create ML sub-models to extract features from photos, and used transfer learning approaches to develop a model that culls sets of photographs based on past samples.
• Utilized pre-trained facial landmark model and tuned EAR hyperparameter, using the mean for multiple faces.
• Achieved model with 90% accuracy in detecting closed eyes. Read about the model on shorturl.at/zDFV1.

• Conducted daily outreach to 30-40 customers, increasing customer engagement by 400% above the daily average.
• Created comprehensive written and video onboarding guides for new clients, which led to a notable increase in customer signups by a factor of two.
• Implemented technical optimizations that significantly improved website performance, including addressing accessibility issues and reducing load times by up to 20%.