RolePark Privacy Policy

1. Who is responsible for your data

The employer company you applied to, which added you to its candidate database, which asked you for a reference or for which you work, is the controller of that data and decides why it is processed. RolePark processes such data as a processor on the company’s behalf under the Data Processing Agreement (https://rolepark.com/dpa).

For the candidate cabinet account, the public talent-pool profile, accounts of company users as customers of the service, billing, the security of the service and analytics of our website, the controller is RolePark (the service operator at rolepark.com).

Contact for personal data questions: help@rolepark.com.

2. What data we process

• Application data: name, contacts (email, phone), photo, CV and details extracted from it (experience, education, skills, languages), expected salary, profile links, screening answers, recruiter comments, interviews, assessments and scorecards with scores, offers, messages in the chat between the candidate and the company, and stage statuses. CVs are parsed on our own server, without external AI services.
• Cabinet account data: email, and — if you sign in with Google — name and profile picture from your Google account; or only the email address if you sign in with a one-time email link; last sign-in date; the CV and photo you upload in the cabinet; jobs you saved and job alert subscriptions.
• Public talent-pool profile: only what you filled in and switched on yourself (position, skills, location, salary expectation, experience summary, optionally name and contacts).
• Referees: if a candidate names you as a referee, the company receives your name, email, relationship to the candidate and your answers to its questions. If you decline, your email is erased and only a one-way code remains so the same company does not ask you again.
• Referred people: if an employee recommends you, the company receives your name, contacts and CV that the employee provided, and the employee’s comment; the employee confirms that you know about the recommendation and do not object.
• Data of company users: name, email, role, photo, actions in the system, messages in the internal chat and in the support chat; if the company uses HR modules — the employee records it enters: position, manager, absences (including sick leave with a supporting document, which is health data), documents, HR Desk requests (including anonymous ones), performance reviews with scores and pulse answers. Salary, personal documents and military registration document data are stored encrypted; military reservation records (obligation, status, dates, reason, note) are stored without field-level encryption.
• Audit and signatures: the company’s audit log records who changed what, with the IP address and browser identifier (user-agent); when a document or offer is signed electronically we keep the time, IP address and user-agent as evidence.
• Payments: for online payment of a plan — company, order, amount and payment status; card details are entered on the payment provider’s page and RolePark does not receive them.
• Technical data: IP address and browser type at the network edge, request logs, cookies and browser storage (see section 9), push-notification subscriptions of company users’ devices, anonymous candidate satisfaction (NPS) answers.
• A company user’s “My notebook”: personal notes (including notes linked to a candidate, vacancy or colleague), the AI assistant’s memory, goals and reminders. In RolePark, personal notes are visible only to their author, and the AI assistant does not read them; AI memory and goals are visible to the author and their AI assistant within the access granted. The company admin and colleagues have no access to any of these records, and they are not part of the company export or the audit log. Reminders are company tasks and are visible like other tasks.

3. Google sign-in, calendars and other integrations

When you sign in with Google we request only basic profile data: email address, name and picture. We use them to authenticate you and, in the candidate cabinet, to find your candidate records in companies that use RolePark by that address. Signing in does not grant calendar access. We do not request access to your Google mail or contacts. The candidate cabinet can also be opened with a one-time link sent to your email instead of Google.

Google Calendar integration is optional and is connected separately by a company user in RolePark Settings with their consent. RolePark creates, updates and deletes interview events in the connected user’s calendar and generates Google Meet links, and reads events from the user’s calendars at the moment the calendar is shown in RolePark, to display busy time and events created outside RolePark; read events are not stored. Event details sent to Google include the interview title, date and time, vacancy title, candidate and interviewer names and email addresses, and a link to the candidate cabinet in RolePark. Google Calendar sends invitations and updates to the selected participants.

To synchronize interviews, RolePark stores the connected Google account email, encrypted access and refresh tokens, connection status information, Google event IDs and meeting links. Disconnecting Google Calendar in Settings removes the saved integration credentials and sends Google a request to revoke authorization. Previously created events are not automatically deleted; you can manage them in Google Calendar. Interview records, event IDs and meeting links remain subject to the retention and deletion rules for the corresponding data in RolePark.

Our use of data received from Google APIs complies with the Google API Services User Data Policy, including the Limited Use requirements. We use this data for the sign-in and calendar features described here, do not sell it and do not use it for advertising.

4. AI assistants: Claude, ChatGPT and other MCP clients

A company user can connect RolePark to their own AI assistant — Claude (Anthropic), ChatGPT (OpenAI) or another client that supports the Model Context Protocol (MCP). The person connects it themselves: they sign in to RolePark and approve access on our consent screen. For self-registered companies, AI assistant access (view and changes) is enabled from registration; for other companies, an administrator enables it. Data is transferred to the AI provider only after the user connects the assistant themselves and approves access on the consent screen. The administrator may choose “off”, “view only” or “view and changes” at any time. For candidate data, the employer company remains the controller (section 1).

What the assistant can see. The assistant works with the access of the person who connected it and never more: only what this person can see in RolePark, within their role, their company and the access granted. With view access (rolepark.read) it can read candidate profiles and applications (including name, contact details if the person can see them, current position, location, salary expectations, experience, education, skills, tags, source and pipeline stage), vacancies and their pipelines, candidate activity, notes and comments, the person’s interviews and tasks, and the person’s own name, email, role and company. Fields hidden from the person stay hidden from the assistant.

What the assistant can change. The AI memory in the person’s My notebook — already with view access (see below). Everything else only with separate change access (rolepark.write), and only if the company allows changes: add a candidate with the source of the data, add a candidate to a vacancy, move an application to another stage, schedule an interview (the application may move forward to the interview’s stage as the company’s stage-order rule allows — under the default rule only when that stage is the next one; the candidate and interviewers receive the usual invitation emails), add a note, create a task, including one for a colleague. Moving a candidate to offer, hired or rejected sends an email to the vacancy’s recruiter and hiring manager (except the person who made the move and anyone who turned these notifications off), not to the candidate; if the candidate still has interviews scheduled, the assistant can cancel them together with the move (otherwise the move is refused), and the candidate then receives the cancellation email. The assistant cannot delete data, change settings or download files; the exception is the AI memory entries in the person’s My notebook (see below).

AI memory. In My notebook, a person can keep AI memory: short entries about their preferences and working rules. The assistant reads it and saves, edits and deletes entries already with view access: it is the person’s own memory, not company data. The person sees and can change every entry, there is no hidden memory, and other people in the company cannot see it. Personal goals in My notebook: with view access the assistant reads them (title, target, progress and deadline); with change access it sets new goals and updates their progress and target, and a goal it sets is marked as written by the assistant. The assistant cannot rename a goal, change its deadline or delete it; in My notebook the person updates the progress or deletes the goal. The assistant does not read the person’s notes in My notebook. The assistant receives memory text as data about the person, not as instructions. The company audit log does not record changes to AI memory and goals: these are the person’s own records.

When and to whom data is sent. RolePark sends data only when the connected assistant calls RolePark while working on the person’s request, and each response contains only the fields needed for that task. The data goes to the provider of the assistant the person chose — for example Anthropic for Claude or OpenAI for ChatGPT — which processes it under the person’s or their organisation’s agreement with that provider and its privacy policy. How the provider stores and uses conversations is governed by its terms, not by RolePark. RolePark has no access to the assistant’s chat history, memory or files.

Purpose. We process this data only so the person can search, summarise and update their recruiting work through the assistant they chose. We do not sell this data and do not use it for advertising.

What RolePark keeps and for how long. For each connection: the application name, the person and company it belongs to, the access granted, and when it was connected and last used. A revoked connection disappears from the list of connections, but the record of it is kept for as long as the company account exists. Tokens are stored only as hashes: an access token is valid for 1 hour and a refresh token for 30 days; expired or revoked tokens are deleted 30 days later. Every change made through the assistant is recorded in the company’s audit log as the person’s action, with the application’s name, the IP address of the request (for an assistant, the address of its provider’s server) and a technical client identifier (user-agent), and is kept for as long as the company account exists. Requests from the assistant also appear in our technical request logs (sections 6 and 7).

How to revoke access. In RolePark: Profile → AI agents → “Disconnect”; the assistant loses access immediately. A company administrator sees all of the company’s connections there and can disconnect any of them, or switch AI agents off for the whole company in Company settings → AI agents. When a user is deactivated, all of their tokens stop working. In the assistant: remove the RolePark connector in its settings (for example Customize → Connectors in Claude or the apps settings in ChatGPT). Removing the connector does not delete conversations the provider has already stored; manage those in the assistant.

Connecting assistants is a feature for company staff at work and is not directed at children (see section 11).

5. Purposes and legal bases

• Reviewing your application and communicating about it — based on the consent you gave when applying, or on the employer’s legitimate interest if the CV arrived another way (for example added by a recruiter, imported or recommended by an employee). The employer, as controller, decides and is responsible for the basis.
• Reference checks — at the request of the candidate, who consents to the company contacting referees, and the company’s legitimate interest in checking the information; a referee may decline or object at any time.
• Candidate cabinet — based on your consent and to provide the features you use.
• Public talent-pool profile — solely on your explicit consent: the profile stays hidden until you switch it on; contacts of an anonymous profile are revealed to a company only after you accept its invitation.
• Accounts of company users and billing — performance of the contract with the company; accounting records — legal obligation.
• Employee records in HR modules, including sick-leave documents (health data) — the employer company decides and is responsible for the basis, for example obligations under employment law.
• Analytics cookies — your consent; Cloudflare Web Analytics without cookies, service security, abuse prevention and legal compliance — legitimate interest.

6. Who we share data with

• Employer companies you applied to, whose invitation from the talent pool you accepted, or which asked you for a reference.
• Infrastructure providers acting on our instructions: hosting of the application server; Cloudflare in front of the website (protection, delivery, analytics without cookies); Google Cloud Storage for files and backups; email delivery through our own mail server, Resend and Brevo; Grafana Cloud (Grafana Labs) for technical logs and monitoring — it receives user and company identifiers, the user’s role, the company name (for a sole proprietor this may be a person’s name), and the address and result of each request, without IP addresses, search queries or request content; browser error reports may contain a technical description of the error. The full list with locations is at https://rolepark.com/subprocessors.
• Google and the selected interview participants — event details, if a company user connected Google Calendar, as described in section 3.
• Zoom and Calendly — only if a company user connected them. Zoom receives the interview title and the candidate’s name to create a meeting. Calendly receives the candidate’s name and email to prefill the booking link; RolePark can also book a slot for the candidate directly (name, email, time zone and the application identifier) and cancel a booking, and Calendly returns booking details to RolePark.
• Telegram — through RolePark’s own bot, to company users who connected it. Approval messages contain the candidate’s name, and workflow steps send the texts the company wrote in its templates, which may include the candidate’s name and email; an employee who connects the bot also sees their own position, manager and leave requests in it.
• Browser push services (Google, Mozilla, Apple, Microsoft) — a short generic notification text without names, if a company user turned push notifications on.
• The provider of an AI assistant that a company user connected themselves (for example Anthropic or OpenAI) — only the data the person requests through the assistant, as described in section 4.
• Systems your employer connects: if the company set up API access, webhooks, single sign-on (SAML), user provisioning (SCIM) or an employer job feed, RolePark sends the relevant data (for example the name, contacts and CV link of an application) to the addresses the company specified. The company is responsible for those systems.
• LiqPay (PrivatBank) — order and payment data when a company pays for a plan online.
• JetBrains YouTrack — messages and screenshots that you send us yourself through “Feedback”, so we can track and fix the issue, and technical error reports.
• Google (Google Analytics 4) — only after your consent and for our own purposes, as described in section 9.
• Microsoft (Microsoft Clarity) — only after your consent and for our own purposes, with all text masked, as described in section 9.
• International transfers: RolePark is operated from Ukraine, and some providers process data in the EU, the United States and other countries. Where the GDPR applies, transfers rely on adequacy decisions or the EU Standard Contractual Clauses (see the Data Processing Agreement).
• Public authorities — only where the law requires it.
• We do not sell personal data and do not use it for advertising.

7. Retention periods

• Application data is kept until the employer company deletes it or deletes its account; RolePark does not delete it automatically, and the company decides how long to keep it.
• Withdrawing consent stops new emails about the application, except reminders about interviews already booked, requests to sign documents and requests to referees that are part of the ongoing process; it does not delete the data by itself.
• To delete the data, delete your cabinet account or ask the company. The “That wasn’t me — delete my data” link in the confirmation email withdraws consent; the candidate record is deleted only if this application created it and nobody has worked on it yet, otherwise the company decides.
• Some copies remain after a candidate record is deleted: the raw data of an import, the original of an inbound email, electronically signed documents with the signer’s evidence, attachments to task replies and events in a connected Google Calendar.
• Reference requests and referees’ answers are deleted together with the application.
• Cabinet account and public profile — until you delete them or ask us to.
• Technical request logs — up to 14 days in Grafana Cloud; database backups — 30 days.
• My notebook — until the person deletes the records; together with the company account or the person’s account; 30 days after the person is disconnected from the company (deactivated or moved to another company). While in the company, the person can download the records at any time on the My notebook page. After disconnection we email a download link to the email address of their RolePark account; the link works for 30 days, until the records are deleted.
• The full schedule, including what remains after deletion, sensitive HR data, logs, sign-in links and surveys, is at https://rolepark.com/data-retention.

8. Your rights

• Access your data, get a copy of it, correct it or delete it; object to processing based on legitimate interest; ask to restrict processing; receive the data you provided in a portable format where this applies.
• Withdraw consent at any time: in the candidate cabinet with the “Revoke consent” button next to an application, or by email to help@rolepark.com. Withdrawal does not affect the lawfulness of processing before it.
• Hide or delete your public profile, decline a company’s invitation, delete your cabinet account in the cabinet (this also deletes the candidate records found by your verified email in all companies that use RolePark, except the copies listed at https://rolepark.com/data-retention).
• Company users: ask your company administrator, or write to help@rolepark.com.
• For data controlled by an employer company we forward your request to that company and help it answer. We respond to requests within the time limits required by applicable law; we may ask you to confirm your identity, for example by replying from the email address in question.
• Lodge a complaint with a data protection authority: in Ukraine — the Ukrainian Parliament Commissioner for Human Rights; in the EU — the authority of your country; in the United Kingdom — the ICO.

9. Cookies

Necessary cookies: sign-in session, interface language and your analytics choice (`rp_consent`, `rp_consent_v`). They always work because the service cannot work without them. The interface theme and similar settings are kept only in your browser storage.

Analytics cookies are set only with your consent. On your first visit we show a banner: until you press “Accept”, Google Analytics is not loaded. After consent, Google Analytics 4 (cookies `_ga` and `_ga_*`) runs only on public informational pages — the home page, product pages, pricing, methodology, calculators, FAQ, the AI-assistants page and these legal documents — to see which of them people read. It never runs inside the company workspace or the candidate cabinet, on the job board, careers and vacancy pages, public profiles, sign-in and registration pages or pages opened with a link from an email, and it receives no search terms. When you go from an informational page to one of those pages, the page reloads without the Google tag. Consent covers analytics only: Google Signals, ad personalisation and advertising consent signals are turned off, and we do not show ads. Cloudflare Web Analytics counts page views without cookies.

With the same consent Microsoft Clarity records how pages are used — layout, clicks, scrolling and the page address — on the public informational pages, the registration and sign-in pages and the first screen of the workspace of companies registered within the last 14 days (cookies `_clck` and `_clsk`). All text and images are masked in your browser and what is typed into form fields is never recorded; Clarity does not run anywhere else. We use the recordings only to find where the service is hard to use.

You can change or withdraw consent at any time with the “Cookie settings” link at the bottom of the legal pages, the home page, the sign-in pages and the public job pages. After withdrawal we delete the Google Analytics and Microsoft Clarity cookies and stop sending data. The full list of cookies and their lifetimes is in the Cookie Policy (https://rolepark.com/cookies).

10. Security

Data is transmitted over HTTPS (TLS 1.2 or higher), sessions are protected with httpOnly cookies, data of each company is isolated from other companies, and files are available only to authenticated users of the company concerned. Integration tokens and sensitive HR fields are encrypted, passwords are stored as hashes, two-factor authentication is available, and repeated failed sign-ins are blocked. Changes are recorded in the company’s audit log. If a breach affects your data, we will inform the company concerned, and you and the authorities where the law requires it.

11. Age

The service is intended for persons aged 16 and over.

12. Changes to this policy

We may update this policy; the current version is always available at https://rolepark.com/privacy with its version and effective date at the top. We will inform users in advance of material changes.

13. Contact

Personal data questions and requests: help@rolepark.com. Operator: RolePark (the service operator at rolepark.com).