Privacy Policy

Last updated: August 21, 2026

The short version

Resume Notebook can be used three ways, and they store your data in genuinely different places. Which one you are in is the single most important fact in this policy, so it comes first.

	Where your data lives	What we know about you
Browsing as a guest	Your browser only	Nothing that identifies you
Trial session	Our database, under a temporary account	Whatever you type or the assistant writes
Registered account	Our database, under your account	The above, plus your email address

You move from the first to the second the moment you start a conversation with the assistant. That is the only thing that does it, and it is described in detail below.

We don't sell your data, and we don't use it to train anybody's model.

If you open the career builder and start typing, your work is saved in your browser's local storage under the key resume-builder-guest. It never leaves your device. We cannot read it, we cannot recover it for you, and clearing your browser data deletes it.

In this mode we hold no record of you beyond the ordinary web-server logs and the analytics described under What else we collect.

Trial session (an automatically created temporary account)

The assistant cannot run in your browser — it needs a server to talk to a language model and to store the conversation. So the first time you use it, we create an account for you automatically, without asking for an email or a password. This happens when you:

send a message from the chat box on the homepage;
choose "fix this with the assistant" after running the free ATS checker; or
send your first assistant message from inside the builder after having used it as a guest — at which point anything already in your browser's local storage is copied up to the new account.

From that point on you are using the database like any other user, and this policy's account sections apply to you. Concretely, the temporary account holds:

Your career data — everything you enter or the assistant writes down: your name, contact details, work history, education, skills, and so on.
Your resumes and the choices you made building them.
Your conversations with the assistant, including its replies and the edits it made on your behalf.

A trial session has no email address and no password. It is tied to the browser that created it, so it can't be recovered, moved to another device, or looked up by us — there is nothing to identify it by.

We delete trial sessions 30 days after they are created, along with the career data, resumes, and conversations they hold. That is a fixed lifetime, not an idle timer: it runs out whether or not you're still using it. Sign up before then if you want to keep any of it. You can also delete it yourself at any time from Settings → Data → Delete Account.

Signing up converts the trial session into a registered account in place. It keeps the same underlying record, so nothing is copied, moved, or lost — it simply gains an email address and a password, and stops being on a timer.

Registered account

A registered account is a trial session plus credentials. In addition to everything above, we store the email address you sign up with, and use it to sign you in, to send password-reset and email-verification codes, and to send one welcome email.

If you buy a Pro subscription, we also store your subscription status and the customer and subscription identifiers our payment provider gives us. We never receive or store card details.

What else we collect

These apply in every mode, including guest browsing.

Usage analytics. We use Umami, a privacy-focused analytics tool, to count page views and a small set of product events (for example, "a resume was created"). It sets no cookies and does not follow you to other sites.
Error reports. We use Sentry to record crashes and errors so we can fix them. A report includes the error itself, your browser and operating system, a short trail of the actions that led up to it, and — if you are signed in — your internal user ID. It is not meant to include the contents of your career data, though an error message can occasionally quote a value that caused it.
Rate-limit records. To stop one person from creating thousands of trial sessions, we keep a one-way cryptographic hash of the IP address that created one. The hash is salted with a secret only our server holds, so it cannot be reversed into an IP address, and the record is deleted after 24 hours.
Server logs. Our hosting provider keeps standard request logs, which include IP addresses, for a short period.

Resume Notebook itself sets no cookies — your sign-in session is kept in your browser's local storage. Our payment provider's checkout runs on their own site and follows their policy.

Resume files you upload

When you upload a PDF to the ATS checker or to the builder's import, the file is read entirely inside your browser. It is not uploaded to us and we never receive a copy.

Two things follow from that, and the second is the one worth reading twice:

If you only run the ATS checker, nothing about your resume reaches our servers.
If you then hand the result to the assistant, we send it a short summary of what was parsed (a few hundred characters at most) so it knows what you are working from — and, because that starts a conversation, it also creates the trial session described above. The full text you imported is saved into that account as your career data.

The assistant, and what leaves our servers

The assistant is powered by Google's Gemini models. To answer you, we send Google the conversation and the parts of your career data and resumes the assistant is working with. Google processes it to generate a response.

This is the one part of Resume Notebook that cannot happen on your own device, and it is the reason using the assistant moves you out of guest mode instead of quietly sending your browser's data somewhere. If you would rather no third party ever saw your career data, use the career builder as a guest and don't start a conversation — every other feature of the builder works without one.

If you connect Resume Notebook to an outside assistant yourself — via an API key or the Claude connector in Settings — then that assistant reads and writes your real career data, and its provider receives whatever it reads. That connection is yours to make and yours to revoke, from the same settings page. API keys and connector links are available on registered accounts only.

Publishing a resume creates a link at /r/… that anyone holding it can open, without signing in. The link contains an unguessable identifier, but it is not otherwise protected — treat it as public. You can unpublish it, reset it to a new identifier, or give it an expiry date at any time.

Who processes your data

Convex — application database and backend
Vercel — web hosting
Google (Gemini) — the assistant's language model
Resend — transactional email (welcome, verification, password reset)
Polar — payments, as merchant of record; they receive your email and billing details when you buy a subscription
Umami — usage analytics
Sentry — error reporting

These providers may process data in the United States and other countries. We don't sell your data to anyone, and we don't provide it to anyone to train a model.

How long we keep things

Guest data — in your browser until you clear it. We never have it.
Trial sessions — deleted 30 days after the session is created, along with the career data, resumes, and conversations they hold.
Registered accounts — kept until you delete the account.
Rate-limit hashes — 24 hours.
Billing records — kept as long as we're required to for tax and accounting.

Deleting your data

Settings → Data → Delete Account removes your account and everything attached to it — career data, resumes, interview history, and any API keys — immediately and permanently. There is no waiting period and no recovery. It works the same way on a trial session.

If you have only ever used guest mode, clearing your browser's site data for Resume Notebook is the equivalent, and Settings → Data → Clear Career Data does the same from inside the app.