Gunnvant focused on improving documentation quality in the Nixtla/statsforecast repository by addressing terminology inconsistencies related to point predictions. Using Python and Jupyter Notebook, Gunnvant systematically replaced references to 'mean' with 'fitted' in docstrings across multiple model classes, ensuring the documentation accurately reflected the model outputs. This targeted correction enhanced clarity for users and reduced the risk of misinterpretation without altering any model behavior or APIs. The work demonstrated attention to detail in technical writing and typo correction, contributing to better maintainability and more precise release notes. Gunnvant’s efforts provided a foundation for clearer communication within the project.