For Institutions

Get more use from the records you already hold.

We help archives, museums, libraries, universities, and public agencies clean up old records, extract their contents, and make them easier to search.

The citations stay attached. The originals stay under your control.

What we can process

Mixed formats are normal.

Most collections are a mix of scans, exports, handwriting, old OCR, and incomplete metadata. We build the processing steps around the material.

Digital collections

Scans, PDFs, and exports

TIFF collections, PDFs, XML, CSV, database exports, old OCR, and legacy repositories.

Paper records

Photographs and handwriting

Phone photos, bound ledgers, handwritten records, directories, newspapers, and field material.

Multiple systems

Disconnected repositories

Records about the same people, places, organizations, and events spread across different systems.

Existing data

Identifiers and manifests

We can also work with identifiers, signed manifests, and approved data exports.

The work

From page image to usable record.

01 · ReceiveBring in the material under the agreed access rules.
02 · ReadRun OCR or handwriting recognition and repair page layout.
03 · OrganizeIdentify documents, people, places, dates, events, and claims.
04 · CiteAttach page and document references to extracted records.
05 · DeliverProvide search, exports, APIs, or a research interface.
Outputs

Clean records

Processed text, document sections, normalized fields, people, events, relationships, citations, and approved exports.

Built-in citations

Evidence stays attached

An extracted claim carries the passage, page, document, and collection reference needed to check it.

Custody and deployment

Choose a setup that fits your rules.

The collection agreement states where processing happens, who holds the source files, and which outputs may be shared.

Institution-controlled

Process inside your environment

Use an institution-controlled or air-gapped setup when policy or collection restrictions require it.

Hosted

Use Kettle-hosted processing

Hosting is available when the agreement covers storage, processing, and delivery.

Source custody

Keep the master files

Signed IDs, manifests, and approved data can move between systems while the master source files stay with the institution.

Use rights

Set the rules in writing

Processing, hosting, publication, and aggregation are separate permissions in the collection agreement.

Who uses the output

The same records can answer different questions.

First

Archives, museums, libraries, and universities

Give staff and researchers better access to the collection with the original context intact.

Also

Research and genealogy

Find people, events, and relationships across scattered historical records.

Also

Property and public records

Support title, property, planning, and administrative research under the collection's use terms.

Also

Government and enterprise

Search legacy records and deliver approved data through exports or APIs.

We start with the institution and its records. Any later research product must follow the collection agreement.

Send a sampleTell us what you have and what you need from it.