Worked on the kortix-ai/suna repository to enhance document download reliability by implementing RFC5987-based Unicode encoding for Content-Disposition headers. This update ensured that filenames containing Unicode characters were correctly handled across major browsers, improving cross-platform compatibility for DOCX, PPTX, and PDF downloads. Addressed an edge case in the sandboxed HTML-to-docx/ppt/pdf conversion workflow that previously caused Unicode filename errors, resulting in fewer download issues and a smoother user experience. Utilized Python and FastAPI for backend development, focusing on secure file download practices, API development, and robust Unicode support to reduce support requests and streamline user workflows.