Financial Document Collection and Annotation Specialist (Collection + Anonymization + JSON Ground Truth)

Labelers will create a ground-truth dataset by pairing each anonymized financial document (PDF/XLS/image) with a corrected, structured JSON output.
The task consists of :
1) Gather ~10 various financial documents (we will provide support)
2) Extract the required client financial information (primarily assets, liabilities, account/holding details, balances, currencies, dates, and identifying attributes)
3) populate our provided JSON schema/template accurately
4) validate JSON formatting and completeness
5) flag/indicate any ambiguities or missing information using tags/notes.

The outcome is a set of document + JSON “correction” pairs used that we will use to benchmark our model performance and detect improvements/regressions over time.