Recently, I explored using Semantic Kernel and OpenAI for more accurate page scanning. By converting HTML to markdown and extracting metatags with XPath, I reduced input size and improved efficiency. OpenAI’s new JSON output feature further allowed me to receive structured data, making the process more reliable.

I was recently working on learning more about using Dataframes and Polyglot Notebooks with C# and needed some sample data to work with. At first I started by looking for some examples online, but I wanted to write a blog about my experience and wanted my own sample of data.