Misha Zanka

After implementation, chunking worked as intended with `gpt-4o-mini`, but it no longer passes tests.

From [discussion on Reddit](https://www.reddit.com/r/webscraping/comments/1f64lbr/comment/llkfnkt/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button). When scraping youtube channel page, Parsera return's no data. The content is lost during page conversion into markdown. Example: - URL - https://www.youtube.com/@loganpaulvlogs/videos - Column names:...