Legal
Privacy Policy
Last updated: 9 November 2026
How CVder collects, uses, and protects your data, in plain language.
1. Who we are
CVder is operated by Evgeny Savochkin, based in Poland. The service is hosted in the EU (Hetzner, Germany).
You can reach us at support@cvder.app for any data-protection inquiry.
2. What data we collect
- Account data: your email address, name, a one-way hash of your password, and the language the app shows you in.
- Resume content: the resume files you upload and the structured data we extract from them so you can edit, tailor, and export them.
- Resume photo: if you add a portrait to your resume, we store the cropped image on our EU servers alongside that resume. We re-encode it on upload, which removes any camera metadata it carried, such as EXIF location. We use it only to show your resume, in the editor, in the PDF or Word file you download, and on a share link if you publish one (see §7). It is never sent to an AI provider and never used to profile you, and we do not copy it to another service: our off-site backup covers the database only. It is included in your data export (see §6), you can remove it at any time from the editor, and it is deleted with the resume, or with your account if you close it.
- Billing data: your Stripe customer ID and subscription metadata (plan, status, renewal date). Full payment data is handled by Stripe. We never see, store, or transmit your card details. For some purchases, Stripe is the merchant of record for the payment itself (see §4).
- Usage data: timestamps of when you use specific features, and audit events used for security and abuse-prevention.
- Technical data: IP address, browser type, and operating system, used for security and fraud prevention, not for analytics or advertising. The one exception: if you allow Marketing cookies, the browser identification your browser sends when you subscribe goes to Meta with the purchase (see §8).
3. How we use your data
- To provide the resume tailoring service, which includes sending the relevant parts of your resume to AI providers (see §4) so they can suggest improvements.
- To send transactional emails: account verification, password reset, billing receipts, and subscription notices. For a purchase where Stripe is merchant of record (see §4), the receipt for that purchase comes from Stripe.
- To send two getting-started emails after you sign up: a welcome, and about two days later one reminder, only if you have not downloaded a resume or upgraded by then. Each has a one-click unsubscribe link. Unsubscribing stops these two, not account emails.
- To improve the product through aggregate, anonymised usage analytics. We do not profile individual users.
- If you allow Marketing cookies, to measure which of our ads on Meta (Facebook, Instagram) and Google lead to sign-ups and purchases (see §8).
- To comply with applicable legal obligations.
4. Third-party processors
We share the minimum necessary data with the following processors:
- Stripe (USA, GDPR-compliant under Standard Contractual Clauses), payment processing and subscription management. For some purchases, Stripe itself processes the payment as merchant of record, through its Link service: in that case the checkout page links Link's Terms and Privacy Policy, which apply to the payment, and Stripe calculates and collects any applicable VAT or sales tax and issues the receipt. Older subscriptions continue under the arrangement described above.
- Resend (may process data outside the EEA, protected by Standard Contractual Clauses), for delivery of account emails: address verification, password reset, account-deletion confirmation, and billing notices, and of the two getting-started emails (see §3).
- Brevo (may process data outside the EEA, protected by Standard Contractual Clauses), for two things. It delivers the emailed report from the free ATS Checker (we send it your email address and the report we generated for you). Separately, if you opt in to product updates, it holds your email address and your consent flag as our marketing contact list.
- Cloudflare (global), for bot protection ("Turnstile") on the free ATS Checker and the waitlist form. The check runs in your browser, so Cloudflare receives your IP address and browser details when the page loads, before you submit anything. Our server sends only the resulting verification token. It does not forward your IP address to Cloudflare.
- Sentry (EU, Germany), for error monitoring, so we can find and fix crashes. We turn off its personal-data collection, and error reports have request bodies, headers and cookies removed before they are sent. With your consent, it also records a sampled replay of browsing sessions. All text and input fields are masked and images are blocked, so a replay shows layout and interaction, never your content. Nothing is recorded before you consent, and withdrawing consent stops it. Recordings are retained by Sentry for a limited period and are not used for any purpose other than fixing defects.
- OpenAI (USA, GDPR-compliant under a Data Processing Agreement), for AI resume tailoring, content suggestions and interview practice (see §4b). Your resume content is sent to this provider for the duration of the request and is not used to train their models under our DPAs. When a resume you upload has no readable text, such as a scanned or photographed PDF or a screenshot or photo you upload or paste. We send an image of it to the same AI provider solely to transcribe the text. For a scanned PDF we render images of its pages. For an uploaded image we re-encode it first, which removes any embedded metadata (such as EXIF location or camera information) before it leaves our server. These images are processed in memory only and are not stored.
- Google (USA, GDPR-compliant under Standard Contractual Clauses), only if you choose to use it. Signing in with Google shares your email address and name with us, and Google sees that you signed in. Exporting a resume or cover letter to Google Docs sends that document to your own Google Drive. The access we request is limited to files you create with CVder. If you delete your account we revoke that access.
- Meta Platforms (Facebook, Instagram) and Google Ads (both may process data in the USA), only if you allow Marketing cookies, to measure which of our ads lead to sign-ups and purchases. What each receives is listed in §8. Both also use this data under their own privacy policies.
- Hetzner (Germany), for cloud hosting and database storage.
- GitHub, Inc. (USA, GDPR-compliant under Standard Contractual Clauses), for encrypted off-site database backups, so your data survives a failure of the server itself. Backups are encrypted with AES-256 before they leave our server, stored in a private repository, and deleted automatically after 14 days. GitHub cannot read their contents.
4a. Connecting your own AI assistant
You can connect an AI assistant of your choice to CVder using MCP (Model Context Protocol). This works the other way round from the AI processing above: we do not send your data to that assistant's company. Your assistant connects to CVder and requests data, and we return it to the assistant you connected, at your request, under an access you grant and can revoke at any time in Profile → Connected agents.
Two things are worth knowing:
- The assistant you connect is your choice and your relationship, not our sub-processor. Whatever you pull into it is then handled under that provider's terms. For example, if you connect a cloud-based assistant, the data you request travels to that company's servers.
- If you use CVder's own tailoring tool through that connection, the tailoring still runs on our servers and your resume and the job description are sent to OpenAI, exactly as they are when you use the app normally.
Separately, a resume you publish with a share link can be fetched by anyone who has that link, including an AI assistant. Published resumes always have email, phone and location removed.
4b. Interview practice
- In a voice session, your audio goes from your browser to OpenAI, for the live conversation and for transcription.
- CVder does not store audio.
- In a text session, the answers you type are sent to OpenAI.
- In a voice session, a hashed account id is sent to OpenAI as a safety identifier.
- OpenAI processes this data as a processor under a signed Data Processing Addendum (DPA).
- CVder deletes the transcript, and the copy of your resume and job text saved for the session, 7 days after your report is ready, and at the latest 14 days after the session ends.
- Your report is kept until you delete the session or your account.
5. Data retention
- Account data: retained until you delete your account.
- Resume content: retained until you delete it from your account.
- Billing records: retained for 7 years to comply with Polish tax law.
- Audit logs: retained for 12 months.
- Job searches: the job title, country and city of a search are kept as anonymous counts with no link to your account, and deleted once nobody has searched them for 90 days.
- Backups: deleted account data may persist in encrypted database backups for up to 14 days, after which it is removed by automatic rotation.
6. Your rights (GDPR)
Under the GDPR you have the right to:
- Access the personal data we hold about you (request at support@cvder.app).
- Correct inaccurate data, via the Profile page or support.
- Delete your data, via Profile → Delete account, or by contacting support.
- Port your data to another service by downloading it yourself from Profile → Export my data, as a ZIP of JSON files plus any resume photos as images. You can also request it from support.
- Withdraw consent for cookies and marketing emails at any time.
- Lodge a complaint with the Polish Office for Personal Data Protection (UODO) at uodo.gov.pl.
7. Public share links
CVder lets you publish any resume as a public read-only page at /cv/<link-id>. This is an opt-in feature. Every resume stays private until you toggle Share from the Editor.
- What is published: your name, job title, candidate summary, experience, education, skills, certifications, languages, interests, links (LinkedIn, portfolio), and the photo on your resume if it has one. Anyone holding the link can see them on the page and in the PDF it offers for download.
- What is never published: your email address, phone number, and physical location are stripped before the public page is served, even if they appear in your resume content.
- How to revoke: toggle Stop sharing in the Editor at any time. The link starts returning "not available" on the next request.
- Indexing: a published page is public. Search engines and AI assistants may find, index and show it, for example when the link is posted somewhere public, and the page carries Schema.org Person micro-data. We do not submit share pages to search engines. After you stop sharing, the link answers "not available" and search engines drop the page over time.
8. Cookies
We use essential cookies, which are strictly necessary to keep you logged in, run the app and remember settings you choose, such as your language. If you read one of our pages in another language and then sign up, sign in, or follow a link from it to another part of the site, such as the app or a guide, an essential cookie remembers that language so the app opens in it. Signing up also saves that language to your account, and while you are signed in the cookie follows the language on your account, on any device; you can change it in your profile. Unless you turn on Marketing (described below), we use no third-party tracking or advertising cookies and no cross-site identifiers. Everything beyond essential cookies waits for your answer in our cookie banner, never before: first-party analytics (see §9), and advertising cookies from Meta and Google if you turn on Marketing. There are two exceptions, and neither sets a cookie or creates an identifier: the anonymous visit count described in §9, and the count of cookie-banner answers described below. You can withdraw consent at any time from the banner, which the "Cookie preferences" link at the foot of every page reopens. If you enable analytics, this also covers masked session recordings as described in §4.
We also count the cookie banner itself, without identifying anyone. When the banner appears, and again when you answer it, our server adds one to a daily total: banners shown, banners accepted and banners declined. Each total is a date and a number. It holds no cookie, no identifier, no IP address and no browser details, so it cannot be linked to you or to any other visit. It is counted whatever you choose, and it is how we see how many people accept analytics. A browser tab sends at most one count for the banner appearing and one for your answer.
Marketing cookies (off unless you turn them on). If you turn on Marketing, we load Meta's pixel and Google's advertising tag so they can measure which of our ads lead to sign-ups and purchases. They set their own cookies in your browser (such as _fbp and _gcl_au) and, like any request to another site, receive your IP address and browser details. We send them three kinds of event: a page view on our public pages (the home page, examples, the free ATS Checker, pricing and teams), a sign-up when you create an account, and a purchase, with its price and currency, when you subscribe. No resume content is sent, and nothing from your browser names you or gives your email address. Google is told not to use this for ad personalisation.
When you subscribe with Marketing on, our server also reports that purchase to Meta directly, so it is counted even when an ad blocker stops the pixel. It sends the price excluding tax, its currency, a purchase reference, the address of our upgrade page, the browser identification your browser sent at checkout, and a one-way hash of your email address, which lets Meta match the purchase to an ad you saw. It sends no name, no resume content and no IP address. To honour your answer, we keep your latest Marketing answer, and when you gave it, with your account. Nothing is sent if you declined, never answered, or have turned Marketing off by the time the purchase is processed.
Turning Marketing off stops all of this at once: the tags send nothing more, and the cookies they set on our site are deleted.
Some details of how you arrived reach us before you answer the banner, as part of requests you make. Each check in the free ATS Checker sends the campaign tag in the link and the referring site's domain, and the record of the check keeps the campaign tag (see §10). Creating an account, by email or with Google, keeps how you first arrived on your account record (see §9).
If you start signing in with Google or LinkedIn, your tab remembers which one you chose, as a single fixed word in session storage that your browser clears when the tab is closed, so that we can offer to try the same provider again if sign-in fails.
9. Product analytics
We collect first-party analytics about how you use CVder. This includes which features you interact with (for example, AI Improve runs, exports, upgrade clicks), broad device information (browser family, desktop vs mobile), and country derived from your IP address. The IP address itself is not stored. We do not share these analytics with any advertising or tracking network: ad measurement is separate, runs only if you allow Marketing cookies, and is described in §8. You may request deletion of your analytics data at any time by contacting us. See the "Your rights" section.
We process browser analytics only with your consent (GDPR Article 6(1)(a)), given through our cookie banner and withdrawable at any time. If you decline, our scripts send no analytics events, no analytics identifier is created or stored in your browser, and no session recording of any kind takes place. If you accept and later withdraw, collection stops. An identifier already stored in your browser is not erased retroactively. You can clear it with your browser's site-data controls, and you can ask us to delete what we hold.
Whatever your cookie choice, opening one of our pages adds one to an anonymous daily visit count. The count is kept per group of pages (for example, the guides or the pricing page) and per source (for example, a search engine, a tagged link, or a direct visit). The page sends its address, the referring site's domain and any campaign tags in its link, with no identifier. The count keeps only those daily totals, with no IP address and nothing that could tell one visitor from another. It sets no cookie. A flag in the tab's session storage, which holds no identifier and is cleared when the tab closes, stops the same visit being counted twice.
When you create an account, we keep with it how you first arrived: the page you landed on and, where there is one, the referring site's domain, the campaign tag in the link, or the click reference that Meta or Google adds to a link from one of our ads. We keep that reference on our side and never send it back to them. Until you consent, the page holds this only in its own memory, never in a cookie or in browser storage.
Separately from browser analytics, our servers keep technical records of actions that actually happen on the service, such as creating an account, a billing event, or running the free ATS Checker. We need these to operate and account for the service, so they are kept whatever your cookie choice is. For signed-out visitors each such record carries a one-off random identifier generated for that single event, which cannot be used to link one visit to another. For signed-in users it carries an identifier derived from your account.
Two more server records are kept whatever your cookie choice is, and neither sets a cookie. When you are signed in, our server records roughly once a day that your account was in use: a record made at your first request to the app that day, with your account identifier and the same broad browser, device and country details as the records above, and nothing about what you did. We use it to count how many people come back. And when you are signed in and we show you the Pro offer, in the upgrade window or on the pricing page, our server records that it was shown, with which of the two it was and, where known, what opened it and which plan and billing period it showed. We use it to see how many people see the offer before they buy. Visitors who are not signed in produce neither record.
When you delete your account, your user identifier is immediately removed from all analytics records. After 30 days, residual device-level identifiers are also removed, leaving only aggregate, non-identifying counts.
10. Free ATS Checker (no account)
You can use our free ATS Checker without creating an account. Because it works differently from the signed-in product, here is exactly what happens to your data when you use it.
- What we collect: the resume text and job-description text you paste in to be scored, a record that the check happened, including your score and, if the link you arrived by had one, its campaign tag. Only if you ask us to email you the full AI fix-list, your email address, whether you opted in to marketing updates, and how you reached us (campaign and referring site).
- Why we use it: the resume and job-description text are processed to generate your ATS score and AI fix-list. Your email is used to deliver the full fix-list you requested and, only if you explicitly opt in, to send occasional launch and product news.
- AI processing: the text you submit is sent to our AI provider (OpenAI, see §4) to analyse it. As with the rest of the product, it is sent only for the duration of the request and is not used to train their models under our DPA.
- Legal basis: we process the text and email you submit to deliver the check and fix-list you requested (GDPR Article 6(1)(b)). Marketing emails are sent strictly on your consent (GDPR Article 6(1)(a)), which you can withdraw at any time.
- Retention: we do not store the resume or job-description text you submit beyond generating your result. It is processed only for the duration of the request (see §4). We do keep one record of the check itself with your score and our processing cost. After 30 days an automatic job removes the identifier and those details, leaving only the fact that a check happened. Your email and consent choice are retained until you unsubscribe or request deletion. This data is stored in our EU database (Hetzner, Germany).
- Your rights: you can access or delete this data, or withdraw marketing consent and unsubscribe, at any time by emailing support@cvder.app The rights described in §6 apply.
11. Chrome extension (CVder Job Clipper)
The CVder Job Clipper extension for Google Chrome saves job postings to your CVder tracker. It connects to your CVder account and talks only to cvder.app.
- What it reads: only the tab you click the extension icon on (or open it on with its keyboard shortcut), at that moment, the page address, title, text and any structured job data, plus text you select or paste into the extension. When it opens, it sends that page to cvder.app, which reads the job from it to show you what it found and whether that job is already in your tracker. Nothing from the page is kept until you click Save or Auto-tailor, which send it again.
- What we keep in your account: the job we find on that page, title, company, location, description, posting date, job type and the page address, as an entry in your tracker. We do not keep the rest of the page. If you choose Auto-tailor, your resume and the job description are processed like any other tailoring request (see §4).
- Connection and install records: when you connect the extension, we store a hash of its connection key (never the key itself) and a random install id, so you can see and disconnect it. On first install the extension sends that install id once, without any account. We keep only a hash of it for up to a year, to count installs.
- Usage events: we record that a save or Auto-tailor happened, with the job site's domain and the posting's language, never the page content, under the analytics rules in §9.
- On your device: the extension keeps its connection key, the install id and the resume you last picked, in Chrome's extension storage.
- What it never does: it does not collect your browsing history, does not read pages you did not click it on, and does not run on pages in the background. We do not sell this data or use it for anything other than your tracker and the documents you ask for.
- Disconnecting and deleting: disconnect from the extension's settings page or from your CVder profile. The connection key stops working at once. Removing the extension from Chrome deletes what it keeps on your device. Deleting a job from your tracker deletes the saved posting, and deleting your account deletes all of them (§6).
12. Changes to this policy
We will notify registered users by email if we make material changes to this policy. Minor wording updates are reflected on this page with a new "Last updated" date.
13. Contact
For all data-protection inquiries, please email support@cvder.app.