Skip to main content
Forms and intake documents bring a different shape: a person identity at the top, then several parallel lists (employment, education, skills, references). The agent fills out the nested structure in one pass.
The same shape covers job applications and KYC intake. Swap the schema’s outer model and the instructions; the File() plumbing and the agent definition do not change.

KYC intake

Identity verification forms add typed fields the downstream system has to accept verbatim (passport numbers, dates of birth, addresses). The schema should be conservative about types: keep IDs as strings to preserve leading zeros and country-specific formats.
For KYC, every field is review-worthy. Combine this schema with the confidence pattern so the downstream queue knows what to send to a compliance reviewer.

Multi-page applications

Job applications often arrive as multi-page PDFs with attachments. File(url=...) handles a single combined PDF. For loose attachments (cover letter, resume, references), run the agent once per attachment, each with the right output_schema, and merge.
For the resume and the references list, two agent.run(...) calls return typed objects. Compose them into an Application in plain Python.

Schema-shape comparison

The agent code is the same across all four. The schema decides the workload.

Next steps

Developer Resources