Chiti focused on enhancing the accuracy and usability of benchmark documentation within the docling-project/docling-eval repository. Over the course of a month, Chiti identified and corrected errors in the FinTabNet and PubTabNet documentation, ensuring that benchmark names and evaluation commands accurately reflected the datasets and processes in use. By leveraging skills in Markdown and shell scripting, Chiti improved the clarity and repeatability of evaluation workflows, reducing potential confusion for users. The work demonstrated careful attention to detail and a methodical approach to documentation-driven development, resulting in more reliable benchmarking practices and better alignment between documentation and actual evaluation procedures.