Introducing s-1: Sievly’s new document reading model
All posts
Case Studies
OAKMERE rebuilds claims intake around document splitting
OAKMERE uses Split and Classify to break mixed claim packets into the right documents before a single field is extracted.
Ines Okafor
·
Customer Engineering
sievly
OAKMERE × Sievly
Claims at OAKMERE arrive as a single file that can hold a dozen different documents: a first notice of loss, photos, invoices, repair estimates and letters. The underwriting systems team needed each one routed before anything else could happen.
Splitting before extracting
With Split, every packet is broken into logical documents at the page level. Classify then labels each one so the right extraction schema runs on it. Nothing is guessed from file names.
Reviewers stay in the loop
Low-confidence pages go to a review queue in Workbench, where adjusters can see the source page next to the result and correct it in place.
What changed
Claims handlers now start their day with structured, routed files instead of a stack of PDFs. The VP of Underwriting Systems describes it as the first intake process the team did not have to babysit.
More from the blog
sievly
Introducing s-1
Announcements
Introducing s-1, our new document reading model
s-1 reads PDFs, scans and spreadsheets the way a careful analyst does, and returns structured data with a citation for every value. It is available in preview today.
sievly
Reading order
Research
Research note: reading order in complex layouts
Multi-column pages, sidebars and callouts confuse simple reading order. We share what we learned teaching s-1 to follow the page the way a person would.
sievly
Quarryline × Sievly
Case Studies
How Quarryline turns research notes into a searchable knowledge base
Quarryline’s data team uses Parse and Extract to turn thousands of analyst notes, filings and spreadsheets into clean, cited records their AI assistant can trust.