* Led and launched project on Modeling Simplification for Accounting Automation with a Senior

Engineer, saving 4000 hours of manual effort across ~200 projects for Business Accounting teams.

• Stood up BFF infrastructure through Cloud Development Kit that connect backend server to front

end applications, fully setting up the VPCs, DNS, API Gateways, Lambdas, pipelines, metrics, alarms. Technologies used: AWS System, Java, React, Redux, Typescript, CDK

• Performed K-Means Clustering on user vectors and computed closest features to characterize user groups

• Generated user behavior graph and performed Random Walk algorithm to select recommendations

• Evaluated users to detect significant interests based on articles they did and did not click with IG and PMI

• Performed variations of K-Means Clustering (Spherical/Bisection) on keywords to detect interest points

Technologies used: Python

• Individually devised automation of environmental sound effects for Xiaoice’s (Microsoft’s AI system) storytelling (now officially implemented) based on keywords, sentence structure, rules, and SVM

• Conducted research on cross-modal information retrieval (image-to-sound, text-to-sound)

Technologies used: Python, MySQL