Amir Kotobi

- Design and develop data pipelines in Python on simulation and experimental datasets to accelarate data analysis, data visualizations, and predictive modeling - developing and maintaining automated solutions for ETL - Improved analysis and modeling of unstructured data by applying NLP, text mining and feature engineering techniques - Implemented unsupervised machine learning e.g. dimension reduction and clustering on high-dimensional structured data

- Design and implement complex queries in SQL and MongoDB resulted in efficient data management and fast fetching of data - Developed Graph Neural Network models with attention mechanisms -improved model's accuracy and precision-recall AUC from 0.85 to 0.9 - Implemented explainability techniques to improve trust on model's predictions