Collaborated cross-functionally to create a robust program automating the generation of database schemas. This involved meticulous design based on specific requirements and rigorous unit testing to ensure accuracy and reliability.Developed a graphical visualization to illustrate transaction analytics, providing supervisors with intuitive insights into complex financial data, which was subsequently presented to stakeholders.Conducted research on Transaction Cost Analysis (TCA) metrics, translating academic findings into practical implementation using Python pandas within Jupyter Notebook. The resulting work was documented and shared on GitHub for public access and reference.

Created a robust image segmentation model tailored for wind turbines, effectively removing background noise from imagesEngineered an efficient image processing pipeline capable of handling large datasets with hundreds of labeled images. Utilized cluster computing to optimize performance.Implemented transfer learning using PyTorch and pre-trained models from ImageNet, integrating multiple architectures and encoders for enhanced feature extraction.Achieved a significant improvement in model accuracy, elevating the Intersection over Union (IoU) score from 0.69 with semi-supervised learning to an impressive 0.946 using transfer learning techniques.

Developed a Python script implementing a Random Forest Model to analyze census characteristics, generating a comprehensive table detailing average vehicle energy usage.Enhanced computational efficiency by scripting intermediate statistical calculations for vehicle energy usage, leading to significantly faster data processing for millions of vehicle records compared to traditional Excel methods.Visualized intricate vehicle statistics using data visualization libraries Matplotlib and Seaborn, aiding my supervisor's decision-making process and research paper quality

Retrieved data from text, pdf and excel files of different fisheries and converted them into a stock-year-quota format in CSV files using tools in Python.

Designed graphs with ggplot2 using data from online databases on frequencies of tornados across the US.Used workflow to set up a site with findings and graphs comparing different years and types of storms.