Loading page
About this tool
Anonymize resumes locally for blind hiring by replacing names, contact details, addresses, dates of birth, links, handles, and identifiers.
The Resume PII Stripper redacts a resume for blind hiring by detecting eight categories of personal data — names, emails, phone numbers, addresses, profile links, social handles, dates of birth and identity numbers — and either deleting them or replacing them with stable placeholders such as [NAME_1] and [EMAIL_2]. The same value always maps to the same token, so a candidate stays traceable through the document without ever being identifiable. It is for recruiters and hiring panels who need anonymised CVs, and for candidates who want to share a work history without handing over contact details.
Open Resume PII Stripper on AltFTool — it loads instantly in your browser.
Paste your CV into the Resume text box, or use Optional local file to open a TXT, DOCX or text-based PDF up to 15 MB.
Untick any of the eight detection categories you want to keep, then pick Stable placeholders (repeats reuse [EMAIL_1]) or Remove detected values.
Check the Anonymized preview, then Copy preview or Download TXT; Download JSON saves the counts-only <name>-pii-counts.json report.
Each distinct value gets a numbered token like [PHONE_1] that is reused everywhere it appears, so the redacted resume still reads coherently.
Beyond patterns like "Name:", it scans the first eight non-empty lines for a 2-to-4-word capitalised line and skips section headings such as Summary, Experience or Skills.
The exportable report is counts-only: detections per category and how many distinct values, with the resume text and the detected values deliberately left out.
Eight categories: names, email addresses, phone numbers, postal addresses, profile and portfolio URLs, @ social handles, dates of birth, and identity or candidate identifiers such as passport, employee ID and national ID numbers. Each category can be toggled off if you want to keep it.
Placeholder mode substitutes a numbered token — [NAME_1], [EMAIL_1], [URL_2] — and reuses that token for every occurrence of the same value, so the document stays readable and internally consistent. Remove mode deletes the text outright and then tidies up trailing punctuation, empty labels and runs of three or more blank lines.
A candidate number must contain 10 to 15 digits, must not be all the same repeated digit, and must either start with + , contain separators like spaces, dots, dashes or brackets, or be a 10-digit number starting 6 to 9. That keeps ordinary numbers such as years, salaries and metrics from being redacted.
No automated detector is complete — an unusual name format, an address with no street keyword, or an identifier written in an unexpected pattern can slip through. Always read the redacted output before sharing it, and treat the detection summary as a checklist rather than a guarantee.