What is Document Blueprint?
Document Blueprint is a config-driven platform that turns document-heavy processes into deterministic pipelines using templates, cases, and AI extraction.
Last updated
Document Blueprint is a config-driven platform that turns document-heavy processes into deterministic pipelines using templates, cases, and AI extraction.
It's built for teams that process the same documents repeatedly — invoices, permits, forms, contracts — where most of the real work is configuration: what data goes where, what outputs get generated, who reviews what. Define that configuration once and the engine runs it the same way on every new document.
Core concepts
Workspaces
A workspace is the top-level container for one document process. It owns a dashboard, an inbox of cases, a set of templates, and a file-category routing table. Teams typically run one workspace per process; switching workspaces changes everything you see.
Templates
A template is the blueprint for a generated document. It defines fields, their data sources, where they land on the page, and how the output file is named. Once configured, autofill produces identical output for every input.
Cases
A case is a unit of work — one invoice, one permit application, one contract. It holds files, extracted data, generated outputs, notes, and activity history. Cases live inside workspaces and inherit their configuration.
AI extraction
When a file is ingested, AI reads it and pulls structured data into fields whose source is extracted. The template's field outputs bound the AI, so extraction stays predictable. Humans review before anything downstream happens.
Walkthrough: invoice processing
A construction company receives 200 vendor invoices per week. Each needs categorizing, extracting into line items, matching against a project, and filing with accounts payable. Without Document Blueprint, that's hours of manual data entry.
With Document Blueprint:
- Invoices arrive via Gmail sync, Drive import, or manual upload.
- An automation routes each file to a category based on sender, filename, or content.
- The matching template extracts fields like
Invoice Number,Vendor Name,Amount Due, andDue Date. - A worker reviews the extracted data in the case detail panel and corrects anything wrong.
- Autofill generates the downstream PDF — an AP cover sheet, a routing form — using the extracted values.
Common patterns
Config-driven workspaces
One engine runs every workspace. The difference between a construction invoice workspace and a legal intake workspace lives entirely in template configuration — change the fields and categories and you've changed the process.
Deterministic generation
Downstream systems — courts, lenders, accountants — often reject documents that look slightly different each run. That's why coordinate placement is exact: when a template writes to coordinates, it writes to those exact coordinates every time.
Mailboxes and filter presets
A mailbox is a persistent partition of the workspace's cases — every workspace gets Inbox, Archived, and Trash, and you can add custom mailboxes for any state worth tracking. A filter preset is a saved set of filter clauses (status=review, due_date < today, assignee=me) that you can apply on top of any mailbox view. Together they let teams triage without touching underlying case data.
Portal collaboration
Share specific cases with specific people outside your team — clients, contractors, plan reviewers. They get a scoped view of only the fields and files you allow, and can fill in whatever you left for them.
Troubleshooting
The AI extracted the wrong value. Open the case, fix the field manually, and save. Manual edits are protected — subsequent extractions skip fields you've edited.
My template output doesn't match the sample PDF. Field values land exactly where you place them — there's no auto-layout. Drag the placement marker on the canvas or type exact x/y values.
A file showed up in the wrong workspace. Automations route files based on category. Check the automation's trigger conditions or the file's assigned category.
I changed the template but existing cases didn't update. Template edits don't retroactively re-extract existing cases. Re-run autofill on an individual case if you need the new logic applied.