De-identification, Anonymization & RedactionPage 1 of 6

FOR PDFS, WORD FILES, TEXT AND SCANNED IMAGES

Anonymize and de-identify
unstructured documents instantly.
Replace PII with synthetic data. Keep the original layout.

Try for free Talk to us

How it worksPage 2 of 6

§ 02 · How it works

Four steps. No setup.

From a sensitive file to one you can share, without changing how the document looks.

Processing modesPage 3 of 6

§ 03 · Processing modes

Pick the one that fits your document

Not sure which? Start with the type of file you have. Each card shows exactly what the output looks like.

BEST FOR SCANNED DOCUMENTS

Redaction Sensitive data is blacked out permanently. Black boxes placed exactly over the text, even on poor scans. The original data is destroyed, not just hidden. Built for FOIA, legal discovery and compliance releases. Scanned PDF · DOCX · Images

BEST FOR TEXT-SEARCHABLE DOCUMENTS

Text replacement Sensitive data is swapped with realistic synthetic data. The same person gets the same synthetic name everywhere, so the document still makes sense. Stays searchable and editable. Useful for AI training, analytics and research. Native PDF · DOCX

BEST FOR TEXT + IMAGE DOCUMENTS

Multiple data types Text is replaced, images are blacked out. For brochures, SOPs and client documents that mix text and visuals. In one pass, Re-Doc replaces text and covers logos, stamps and signatures inside images. DOCX · Logos · Signatures

Full walkthrough

See all three modes in one demo

What our customers sayPage 4 of 6

§ 04 · What our customers say

TS AK RN +

More than 100 users have tried Re-Doc

10,000+ pages made safe to share

500+ documents processed

95 to 99% of sensitive data detected correctly

§ 04.1 · Customer stories

What our customers say

Teams handling sensitive documents use Re-Doc to make them safe to share.

From: Hub.xyz · YC S26About: a document data project

“Re-Doc was very helpful for a specific data project we ran for a frontier lab. We usually work with image, video and audio data, so documents were new territory for us, and Re-Doc helped us tackle it.”

Tim SprecherFounder, Hub.xyz

API accessPage 5 of 6

§ 05 · API access

Processing thousands of documents?

Our REST API handles batch redaction and text replacement at scale. Plug it into the document workflows you already run.

Read the API documentation

STEPENDPOINTSTATUS

01 Redact scansPOST /api/v1/processing/upload Scanned PDF, image or DOCX, up to 50 MB. Black boxes go over the detected data. ● Available

02 Replace textPOST /api/v1/text-replacement/process-searchable-pdf Native PDF, DOCX or TXT. Synthetic data replaces the real values and the layout stays. ● Available

03 DownloadGET /api/v1/processing/download/{id}/converted/0 Get the safe file back in its original format. Text replacement results download from /text-replacement/download. ● Available

§ 05.1 · Security

Your documents stay yours

Never used for training Documents are never used to train models or stored beyond what processing your file requires.

ZDR-approved LLM gateways The LLM gateways behind Re-Doc are Zero Data Retention approved.

Deleted on your schedule Documents are processed and deleted from our servers within 7 to 30 days, depending on what you request.

Get startedPage 6 of 6

Still not sure if this is enough for you?

Talk to us about your documents and your volume, or ask for a paid proof of concept run on your own files.

Talk to us Request a paid POC

Anonymize and de-identifyAnonymizeDe-identifyunstructured documents instantly.Replace PII with synthetic data. Keep the original layout.Replace PII with synthetic data.Keep the original layout.Keep it safe to share.