ResumeReadoutWe wrote this in plain English on purpose. If anything here is unclear, that’s a bug in the policy, not a reason to keep reading between the lines.
What happens when you check a resume
Reading your file — checking the text layer, detecting columns and tables, pulling out contact info, matching section headers — runs entirely in your browser, in JavaScript executing on your own device.
The file itself is never transmitted to our servers. We don’t receive a copy, we don’t store one, and we couldn’t produce one if asked.
The optional AI keyword match
If you choose to paste in a job description to get keyword suggestions, we send the text extracted from your resume — not the file — along with that job description to our server, which forwards it to Anthropic’s Claude API to generate suggestions.
- That text is used only to generate the response for that one request.
- We don’t store it, log it long-term, or use it to train anything.
- This step is entirely optional. Skip it, and no resume text reaches our servers through this feature — see below for the one other thing that can, if you choose to use it.
The optional accuracy feedback
If you click “Was this accurate?” on your results, we record which checks passed or failed for that scan (for example, “contact info: not recognized”) and, if you choose to add a note, whatever you type there.
- We never see your resume’s text or file through this feature — only which named checks returned which result, plus your optional note.
- Unlike the AI keyword match above, this feedback is stored — that’s the point, it’s how we find and fix real accuracy problems. It isn’t tied to your name, email, or any other identifier.
- This step is entirely optional. If you don’t click anything, nothing is sent.
Accounts and personal data
There’s no account, no signup, and no email required to use the free parse-check. We don’t ask for or store your name, email, or any other personal identifier tied to your resume.
Analytics and error monitoring
We use Vercel Analytics for aggregated, privacy-friendly page-view stats. It doesn’t use cookies and doesn’t track you individually across sites.
We use Google Analytics for traffic and usage stats. It sets a cookie to distinguish visitors and sends standard analytics data (pages viewed, general location from IP, device/browser type) to Google. We haven’t configured it to read anything you enter into the page — just how the site as a whole gets used.
We use Sentry to catch and fix bugs. For a sample of visits — and for any visit where something breaks — it can capture a masked replay of the page so we can see what went wrong. Text and input content are hidden by default; this is about catching interface bugs, not reading what you typed.
None of these tools ever see the contents of your resume: that stays in your browser, or, for the optional AI feature, goes only to Anthropic as described above.
Who we share data with
- Vercel — hosting and the analytics above.
- Google — analytics, described above. Never sees your resume text or file.
- Sentry — error monitoring, described above.
- Anthropic — extracted resume text and job-description text, only if you use the optional keyword-match feature.
- Upstash — stores accuracy feedback (see above) and powers rate limiting for the AI feature. Never sees your resume text or file.
We don’t sell data, and we don’t have advertisers to share it with.
Cookies
We don’t set advertising cookies, and we don’t use cookies to track you individually across other sites. Google Analytics does set a first-party analytics cookie to distinguish visitors on this site, as described above. Your light/dark theme preference is saved in your browser’s local storage, not a cookie, and never leaves your device.
Changes to this policy
If this changes, we’ll update this page and the date at the top — no silent rewrites.
Questions
Reach us at peachwebhq@gmail.com.