Extracting data from vehicle titles
Each state's motor vehicle agency prints its own certificate of title, with its own layout and security paper. A lender collects one for every vehicle it finances to record its lien, and an insurer collects one to settle each total loss. Titles arrive as posted originals or portal scans.
Title intake needs the VIN, the title number, the owner, the lienholder and any brand such as salvage or rebuilt. A collateral management system matches the VIN to the loan and records the lien. A claims system checks ownership before a total loss is paid.
What makes vehicle titles hard
Every state prints its own title
Box order, section lettering and brand wording differ by state. Any vendor solution relying on position-based extraction needs a template for each one.
Security printing behind every value
Titles print on guilloche borders and patterned fields to deter forgery, and the pattern runs behind every typed value on the face.
Handwritten transfers on the back
Sellers and dealers fill in each assignment and reassignment by hand, with a sale date and odometer reading. Blank sections follow the last completed one.
Parse, extract, verify
Parse
Parse turns the document into reading-order Markdown and a hierarchy of blocks (text, tables, figures), and grounds each block to its exact location on the page.
Extract
You define the fields in a JSON schema, and Extract returns their values in a consistent, structured format, each linked back to where it appears in the Markdown.
Verify
Each value links back to where it appears, down to the character range in the Markdown, and the page and bounding box in the source document, so every result can be verified.
One vehicle title, end to end
About this document: A Commonwealth of Virginia certificate of title for a Ford pickup, two portrait pages scanned from paper. The face carries the vehicle description and a bank lienholder with a signed lien release, then the owning dealership's assignment to a buying dealer. The back holds a completed dealer reassignment, and a FOR EXPORT ONLY stamp crosses both pages. Page 1 is featured. Personal names and identifiers were replaced before publication.
Processing objective: The objective is to identify the vehicle and the title record, the values a lender or insurer checks first on the face. The title printer set all five in one ruled band. The title number prints near a red security control number, and extraction keeps the two apart, each grounded to its own box.
Demonstration fields: VIN, Model year, Make, Title number, Date issued
The document
2 of 2 pages
Page 1shown below
Page 2
Where each value came from
vehicle.vin
{
"value": "1FTEW1E86HFB56934",
"page": 1,
"range": {
"end": 434,
"start": 417
},
"box": {
"xmin": 0.04623,
"ymin": 0.24195,
"xmax": 0.29835,
"ymax": 0.25913
}
}vehicle.year
{
"value": "2017",
"page": 1,
"range": {
"end": 456,
"start": 452
},
"box": {
"xmin": 0.3935,
"ymin": 0.24187,
"xmax": 0.46394,
"ymax": 0.25893
}
}vehicle.make
{
"value": "FORD",
"page": 1,
"range": {
"end": 469,
"start": 465
},
"box": {
"xmin": 0.4637,
"ymin": 0.24183,
"xmax": 0.55669,
"ymax": 0.25889
}
}title_number
{
"value": "9440212714",
"page": 1,
"range": {
"end": 503,
"start": 493
},
"box": {
"xmin": 0.79524,
"ymin": 0.24166,
"xmax": 0.95244,
"ymax": 0.2587
}
}date_issued
{
"value": "07/08/17",
"page": 1,
"range": {
"end": 780,
"start": 772
},
"box": {
"xmin": 0.79575,
"ymin": 0.27019,
"xmax": 0.95242,
"ymax": 0.29
}
}How to try it yourself
Everything on this page is reproducible. The sample, its schema, and the Parse and Extract API outputs are published. You can repeat the same two calls from whichever surface you already build on.
- Claude Skilldescribe the job; the agent writes it
- CLIno code at all
- MCPexpose ADE docs as an agent tool
- REST APIany language
- Pythonlandingai-ade
- TypeScriptlandingai-ade
Documents into data
Accurate
The best accuracy on the difficult documents other tools get wrong, such as messy scans, complicated tables, checkboxes and handwriting.
Auditable
Every answer cites its sources. Values come back with an audit trail (page, bounding box, character range) so a human can review them.
Agentic
Applies multi-step reasoning, and plans before it acts. Handles documents it has never seen before and adapts to variations.
API-first
Composable APIs that drop into any tech stack and can be called from anywhere. Easy to build on and embed in your pipelines and products.
Affordable
Cost-competitive rates with complete price transparency. Spend controls before you run, and every job reports its exact cost.
Enterprise-grade
Built for production and scale from day one. Fully observable and monitored, running securely in our cloud or yours.
Built for high-volume, high-throughput production
- Zero data retention
- Priority rate limits
- Enhanced support
- VPC and on-prem deployment
Available on Team and Enterprise plans. VPC and on-prem deployment require an Enterprise plan.
