Nor Product details scraping, You can do it via JSON, Python, or LlamaIndex If you already know LlamaIndex inside out, great—lean on its APIs for parsing, vector indexing, and fast retrieval. If you’re stronger in raw Python, that’s fine too: build the exact same logic with standard libraries and keep the interfaces JSON-based so I can swap in LlamaIndex later. I’ll supply sample PDFs, CSVs, a target site list for scraping, and a short spec of the queries I need to run. Write clean, well-commented code, include a short README, and demonstrate the pipeline on the sample data so I can run it locally with a single command.