← Back to the screener

Privacy

Your documents are used only for the current request.

Résumé and job-description content is sent only to the configured Resume Keyword Screener API for the current request. The application has no document database or persistent disk, does not intentionally retain the content, and does not transmit it to AI providers or third-party résumé-processing services.

Multipart infrastructure may temporarily spool larger uploads while a request is active. The API bounds the complete request before document parsing and closes framework-managed upload resources after processing. Hosting-provider operational metadata remains governed by the providers' own policies.

This Version 2 frontend is configured to use Vercel Web Analytics for anonymous aggregate page views, referrers, approximate region, browser, operating system, and device category. Analytics receives only an allowlisted public page path—including the optional premium-interest experiment page when enabled—never document text, filenames, extracted terms, report contents, or form input. The app does not use analytics cookies or custom interaction events.

REQUEST LIFECYCLE

What moves—and what remains.

  1. BrowserYou explicitly submit files or text.
  2. API memoryBounded inputs are processed for the request.
  3. Deterministic engineNo external AI service receives the content.
  4. ResultThe response returns without application persistence.

Exports and temporary browser resources

PDF reports and Markdown checklists are created only when you request them. Temporary browser download URLs are revoked after use; downloaded files remain under your browser and device controls.

Document X-Ray disclosure

Contract v2 returns exact canonical document-view text to this tab for X-Ray. It is not anonymized and can be copied or captured by the user. The response is marked no-store and application code does not place this text in analytics, URLs, logs, or browser storage. Browser-memory clearing is not secure erasure.

Resume Lab session comparison

Resume Lab can retain up to five successful result view models and three résumé variants for one job description in the current page's React memory. The application does not write runs, files, source text, review decisions, notes, or comparisons to localStorage, sessionStorage, IndexedDB, URLs, analytics, or a server history. New analysis, Clear Resume Lab, demo clearing, refresh, and tab close release those application references. This is application-state clearing, not a guarantee of secure browser memory erasure.

Temporary revision copies

The Revision Workspace edits a temporary text copy in current React memory. It does not modify an uploaded PDF or DOCX, autosave, or submit while you type. Run Revision explicitly sends that text through the same no-store analysis request. Reset, Clear Resume Lab, New analysis, refresh, or tab close releases application references without claiming secure browser-memory erasure.

Infrastructure metadata

Hosting providers may process ordinary operational metadata under their own policies. The application does not intentionally log document contents or send them to analytics.