AI Parser Training

I have a batch of 12–20 property insurance proposal PDFs—an even mix of ACORD and Indian-insurer templates. I will supply ready-made dummy values for each policy. Your job is twofold:

• Populate every PDF with the data I provide, keeping field placement and formatting intact.
• In Nanonets, create a functional parsing model that reliably pulls 15 target fields (Insured_Name, Risk_Address, Year_Built, Sum_Insured_Building, etc.). Use the completed forms as training material, then verify output against my reference sheet and fine-tune until each field lands in the correct column.

I expect a neat hand-off package that includes:
1. The filled PDFs.
2. The trained Nanonets model (export link or credentials).
3. A concise QA log highlighting mismatches you encountered and how you resolved them.

Everything starts and ends in PDF, so no format conversion surprises. If you already know ACORD layout quirks or Indian proposal jargon, you’ll fly through this. Accuracy and tidy organization matter more to me than sheer speed, but I’d like to wrap this within a week.