• Develop an EDM recommendation model (Bias SVD + Xgboost), improving 10% EDM open rate. • Constructing Salesforce KPI and operation dashboards helped me awarded 2023 best employee. • The ink forecast model assists product managers in controlling ink procurement. • Utilize the collage model to assemble a new image from company picture materials based on a description. • Dump printing volume data from AWS S3 and apply Teams to monitor the data pipeline.

• Evaluated AI routing algorithm and provided advice to the cooperative company on how to improve could achieve the goal we want. • Adopted Xgboost to calculate the value of off-line scooters, generated exchange order organized workers which off-line scooter should be exchanged battery first. • Built LSTM model to forecast rental demand through user app events understanding the fulfill rate of every operational region. • Assessed parking lot value, consulted which one should be deprecated, where could develop new parking lot service. • Applied Airflow to ensure the stability and accuracy of data.

• Crawled articles from different webs and constructed NLP model integrated text-CNN, Logistic, LightGBM, accuracy around 91%. • Analyzed content of articles to calculate content similarity, then recommended readers related articles to improve the click rate and pageview. • Interfaced with GA, dumped data to BigQuery then visualized data via Data Studio to help company partners grasping website traffic. • Advertising optimization, offered the performance and bidding situation of different kinds of advertises to the business department to sell ad slots.

• Developed user segmentation to build CRM helping client administering their customers. • Implemented Xgboost to find the most possible purchase customer list and used Cosine Similarity to recommend their interesting product category causing the repurchase rate 4% improved. • Built exchange rate API compared the current exchange rate with past 30 days then demonstrated exchange recommend rank to users.