Extract data from I-9 forms
Unstract reads Form I-9, Section 1, Section 2 and Supplement B, scanned or e-signed, checks that each one is complete and consistent, and sends the fields to your HRIS or payroll system.
docs seen in person, copies in file — GO
Form I-9 · page 1 of 1 · scanned from packet
{ "last_name": "Lindqvist", "first_name": "Teodora", "date_of_birth": "04/19/1993", "ssn": "***-**-4417", "citizenship_status": "U.S. citizen", "list_b_document": "Driver's license", "list_b_expiration": "08/14/2030", "list_c_document": "Social Security card", "first_day_of_employment": "11/03/2026", "employer": "Pinecrest Logistics LLC", "section_2_within_rule": true}
Trusted by HR and payroll teams around the globe
Every new hire's I-9 reaches HR in a different state
An e-signed I-9 from an onboarding portal is tidy. The forms in a hiring season's file cabinet are not.
Signed at a job site, scanned, then forwarded as a grey copy
Older editions still sit in the file for long-tenured employees
Section 2 filled in by hand, an expiration date crossed out
I-9, ID copies and the background check authorization in one file
Audit lists of hundreds of forms, plus rehire reverifications
HR coordinators check each form by eye and key it into the employee record, and a missed blank or an expired document surfaces only when an audit pulls the file.
Unstract reads every I-9, flags blanks, mismatched dates and expiring documents, and sends the checked fields to your HRIS or payroll system.
Read the I-9, check each section, file the record
LLMWhisperer reads the form, Document Studio returns the fields and the checks, and Flow runs onboarding around them. Each product works on its own, so a team can start with the I-9 alone and add the rest later.
Read I-9 forms with every box, tick and signature date in place
An I-9 is a grid of small boxes: name parts, an apartment number, a citizenship status box, and two document columns in Section 2. Read as plain text, the List B number drifts into the List C column. LLMWhisperer keeps each box where it sits, reports which status box is ticked and which are empty, and reads handwritten dates on scans and phone photos.
Use the text with your own LLM, or make it the first step of a full onboarding run.
Learn more about LLMWhispererTurn each I-9 form into one checked record for the employee file
Describe the record once: Section 1 identity and status, the Section 2 documents with their numbers and expirations, the first day of employment and the employer certification. The I-9 form parser returns the same JSON for a scanned paper form, an e-signed PDF and a Supplement B rehire page. Each record carries checks you set: Section 1 complete, Section 2 signed within the deadline in your rules, no document expired on the first day, and the same name and date of birth as the offer letter.
Deploy it as an API behind your onboarding portal, or as an ETL pipeline over a folder of scanned forms for an internal audit.
Learn more about Document StudioRun employee onboarding from the paperwork to the first payroll
Describe your I-9 and new-hire paperwork in plain language, and Flow's agents build the onboarding workflow from it. You approve the spec and it runs the same way for every hire. Offer letters, I-9s, W-4s and deposit forms arrive from the onboarding portal or the HR inbox, are read and compared, and the checked fields go to your HRIS or payroll system, such as Workday or ADP, with benefits elections passed to benefits administration.
A missing form, a name that differs between documents, an expired I-9 document or a deadline at risk stops for an HR coordinator. Unstract does not decide whether anyone may work, does not run E-Verify and does not examine original documents; those stay with the employer.
Learn more about FlowRead your own I-9s today. Bring onboarding to a consultation.
Start for free to extract fields from a folder of I-9 forms. An onboarding run built around your HRIS starts with a conversation.
Form I-9 and the paperwork that travels with it
Documents
- Form I-9, Section 1 and Section 2
- Supplement A, preparer and translator
- Supplement B, reverification and rehire
- E-signed I-9s from onboarding portals
- Scanned and photographed paper I-9s
- Copies of List A, B and C documents
- Background check authorization forms
- W-4 and state withholding forms
- Direct deposit forms
- Any other new-hire paperwork
Fields captured
For the teams that keep the I-9 file in order
Every I-9 checked for blanks and expired documents as it arrives, and a reverification list built from the expiration dates on file.
New hires reach payroll with a name, SSN and start date that match the I-9, so the first paycheck goes out on time.
I-9s for every client's workforce read the same way, with each record filed under the right client and employer of record.
Seasonal and rehired workers onboarded in batches, with Supplement B pages matched to the original form for each rehire.
Fewer blanks found in audits, fewer forms keyed by hand
Identity documents stay locked down
An I-9 file holds SSNs, dates of birth and identity document numbers for every person the company has hired, the most sensitive file in HR. One leaked folder exposes the whole workforce to identity theft.
- I-9 forms and document copies are encrypted in transit and at rest.
- Employee documents never train any model.
- SSNs and document numbers can be returned masked, and access is set by role, location and client.
- Retention you set, so forms are deleted when your retention rule for former employees runs out.
- Employers that must keep identity data on their own servers can run Unstract self-hosted.
What HR teams ask before they send an I-9
It reads a completed Form I-9 and returns each box as a field: name, date of birth, citizenship status, the Section 2 documents with their numbers and expirations, and the employer certification. HR teams use it to extract data from I-9 forms into the employee record without retyping, and to catch blanks before an audit does.
Unstract reads e-signed and scanned I-9s together with W-4s, state withholding forms and direct deposit forms, and returns each as structured data. The AI data extraction runs in Document Studio, and Flow carries the checked fields into your HRIS or payroll system.
Yes, for the paperwork. Unstract reads the form a remote hire sends, checks that Section 1 and Section 2 are complete and that the dates fit your rules, and flags what is missing. The employer still examines the identity documents in the way its own procedures allow.
Yes. Run the extraction as an ETL pipeline over a folder of scanned forms and every I-9 comes back with the same fields and checks: missing boxes, expired documents, late Section 2 dates under your rules, and work authorizations due for reverification. HR then reviews the exceptions, not every page.
No. Unstract reads what the form says and checks it for completeness and consistency. Eligibility decisions, E-Verify cases and the examination of original documents stay with the employer.
Yes. Background check authorization form OCR reads the candidate's name, date of birth, consent box and signature date, typed or by hand, and the record shows whether consent was signed before the check was ordered.
Yes. The new name, rehire date and reverified document with its number and expiration come back as fields, linked to the employee's original Section 1 so the history reads in one record.
The field comes back empty and flagged, never guessed, with the spot on the page it came from, so an HR coordinator can correct it in seconds.
Unstract is SOC 2, ISO 27001, GDPR and HIPAA compliant. Forms are encrypted, never used for training, and deleted on the schedule you set; self-hosting is available. Your security team can review each control in the Trust Center.
Not to try it: start for free and upload a few I-9s. For onboarding connected to your HRIS, we build the run with you in a consultation.
Bring a drawer of I-9 forms
Paper forms, e-signed PDFs, Supplement B pages and a few background check authorizations. We read them in the session and show the record each one produces, with every blank and expired document flagged.