Engineer streaming processing data pipeline on Azure with the main purpose to ingest and process tweets and satellite images data from Hurricane Harvey natural disaster, and serve Power BI.

Built a deep learning model to analyze and predict hurricane patterns using meteorological data. Applied data preprocessing, feature engineering, and neural networks to identify trends and improve forecasting accuracy, aiding disaster preparedness and risk management.

Hurricane Track Prediction using a probabilistic Kalman Filter on 183 years of IBTrACS data, featuring full EDA, feature engineering, and open-loop forecasting evaluation across 13,450 storms structure Implements a feature-adaptive state-space model with data-driven Q/R estimation.