User Story:
As a PSAP engineer,

I want a in-house evaluation dataset,

So that we can validate the RAG and fine-tuning methodologies we use in performance testing.

Acceptance criteria:

• A dataset of PDFs converted to Markdown with clear guidelines on data cleaning and format.
• QNA files derived from the dataset for use in the RHEL AI SDG process.
• A set of questions for benchmarking.