- Built and deployed end-to-end data pipelines using Python and SQL, employing technologies such as SageMaker Pipeline, Lambda, and PySpark to process finance data and event data with billions of rows efficiently.- Designed and implemented fine-tuned anomaly detection models and dashboards for financial data forecasting pipelines, significantly improving financial forecast accuracy and earning recognition from both the Principal PM and DS Manager.- Engineered and optimized OLTP and OLAP data models, providing robust support for critical financial operations.- Designed and implemented scalable data pipelines integrating CI/CD best practices to ensure high reliability and operational efficiency.

Azure Sphere

1. Disaster Response Web AppAnalyzed disaster data form Figure Eight to build a data pipeline and a machine learning pipeline to categorize emergency messages so that an appropriate disaster relief agency can reach out and help.The project includes a web app where can quickly identify a message's categroies. It also visualized the messages. - Built an ETL data pipeline to prepare message data - Built a machine learning pipeline based on BOW, Mult Output Classifier, and Grid Search - Build a web application that uses Flask, Plotly, and the Bootstrap framework2. Image Classification- Built a Convolutional Neural Networks image classification model to identify species of flowers- Processed image dataset and set up image generators to feed the data to train the model - Developed an end to end CLI tool based on the trained model3. Starbucks customer obtain and retain projectUsed purchasing habits to identify groups of individuals that are most likely to be responsive to rebates and discount