CareerMate Privacy Policy
Last updated: 19 June 2026 Effective date: 19 June 2026
This Privacy Policy explains what personal data we collect, why, and what your rights are. It is incorporated by reference into the CareerMate Terms of Service.
We take your privacy seriously. CareerMate is an interview-practice tool — a preparation aid you use before a real interview. We don't sell data. We don't show ads. We don't track you across apps or sites. We do not profile your voice, score your delivery for hireability, or build a "skill" profile of you across sessions.
1. Data Controller
The data controller for personal data processed by CareerMate is:
Ilia Vlasov, operating as "w4app" (sole trader), domiciled in Finland. Contact: privacy@w4app.com (or support@w4app.com)
We plan to transfer the Service to w4app Oy in 2026 (expected Q3 2026). When that happens we will notify you and reissue this Policy under the new controller's name without reducing your protection — see Section 14.
We do not currently have a designated Data Protection Officer (DPO), as appointment is not mandatory under GDPR Art. 37 for an entity of our size and scope. We monitor this threshold; if our processing activities meet the DPO appointment criteria, we will appoint and disclose one.
2. What Data We Collect
Plain summary: your account identifier, your purchase/subscription status, the interview answers you type or speak (saved as text and session summaries on our servers), and — only if you opt into voice answers — the audio of your spoken answer, which is transcribed and then deleted. That's it. No contacts, no health data, no photo library, no advertising identifiers, no voice-print, no emotion analysis.
2.1 Account Data
- w4app / CareerMate account identifier (created via magic-link sign-in)
- Email address (used for magic-link sign-in and account communications)
- Subscription / entitlement status (from Apple via RevenueCat), including the active product and free-tier usage state
- Push-notification device token, if you enable notifications (for delivery via Apple Push Notification service)
2.2 Interview Session Content (text)
When you run a practice session, CareerMate saves:
- The target role, track, and difficulty you select
- Your typed answers, or the text transcript of your spoken answers (see Section 2.3)
- Session summaries: the standalone sample model answers shown for reference and objective delivery metadata (turn length, pause length, filler-word counts)
This content is stored on our servers (PostgreSQL on w4app's infrastructure — see Section 2.5) so you can review and save your past practice sessions. It is your content, used to operate the practice experience for you. We do not use it to score, rank, or characterize you, and we do not condition future content on your prior-session history.
2.3 Voice Answers (optional — special-category handling)
CareerMate lets you answer interview questions by voice instead of typing. Voice answering is optional: you can use CareerMate fully by typing your answers. Voice is enabled only after you give explicit, granular consent (see Section 3, legal basis GDPR Art. 9(2)(a)).
If you choose to answer by voice:
- The audio recording of your answer is sent to our transcription processor (OpenAI, in the United States — see Section 4) for the sole purpose of converting your speech to text.
- We send only the audio recording itself — we do not send your name, email, IP address, device identifier, or any account or session identifier with it.
- Your audio recording is not retained. It is deleted automatically after transcription (on both success and failure), and only the resulting text and a tamper-evidence provenance record are kept.
- We do not use your voice to identify you, and we do not analyse your voice for emotion, tone, sentiment, or personality. No voice-print or speaker model is ever built.
Why this is special-category data. The words you speak in an interview answer can incidentally reveal special-category information under GDPR Art. 9 (for example, health, disability, religion, political opinion, trade-union membership, sexual orientation, or racial/ethnic origin). We do not collect this for that purpose — the purpose is transcription only — but because such content may be present, voice answering is gated behind explicit Art. 9(2)(a) consent and the minimization measures above.
2.4 AI Interviewer Voice (synthetic audio we generate)
The AI interviewer speaks its questions to you in a synthetic, AI-generated voice produced by a text-to-speech processor (OpenAI, see Section 4). This synthetic audio is machine-marked as AI-generated in compliance with the EU AI Act Article 50(2) provenance-marking obligation. The synthetic voice is a neutral interviewer persona; it does not impersonate any real person.
2.5 Diagnostic and Infrastructure Data
- Technical crash and performance diagnostics (device model, OS version, stack traces) to keep the app stable. These are generic on-device diagnostics; we do not send them to any third-party crash- or error-monitoring service.
- App usage events, without content of your answers
Hosting: production runs on w4app's own server infrastructure (Amazon EC2) in the eu-north-1 (Stockholm) region, within the European Economic Area (EEA). Your account and session data are stored in the EEA; no additional international-transfer safeguard is needed for hosting.
2.6 Payment Data
We do not collect or store payment card data. Apple processes all payments via In-App Purchase. We receive only subscription/entitlement status from Apple and RevenueCat.
2.7 What We Do NOT Collect
We do not collect: contacts, health & fitness data from HealthKit, messages or calendar, your photo library, browsing or search history, location, advertising or device-tracking identifiers (no IDFA), or a voice-print/biometric template. We do not use App Tracking Transparency (ATT) because the app does not track. We do not perform emotion, sentiment, or personality inference on your voice or answers.
3. How We Use Your Data
| Data | Purpose | Legal basis (GDPR) |
|---|---|---|
| Account data | Authenticate (magic-link), support, subscription management | Art. 6(1)(b) contract |
| Interview session content (text) | Operate the practice session; let you review/save past sessions | Art. 6(1)(b) contract |
| Voice answer audio (optional) | Transcribe your spoken answer to text | Art. 6(1)(b) contract + Art. 9(2)(a) explicit consent (for incidental special-category content) |
| Synthetic interviewer voice | Deliver the AI interviewer's questions to you | Art. 6(1)(b) contract |
| Push device token (if enabled) | Deliver notifications you have enabled | Art. 6(1)(b) contract / Art. 6(1)(a) consent |
| Diagnostics / usage | Keep the Service stable, fix bugs | Art. 6(1)(f) legitimate interest |
| Payment / subscription status | Operate the subscription and free-tier limit | Art. 6(1)(b) contract + Art. 6(1)(c) legal obligation |
We do not use your data for:
- Selling to third parties or data brokers
- Advertising or ad targeting
- Cross-app or cross-site tracking
- Training AI models (we invoke "no-training" provisions with our AI processor where available)
- Voice-identity / voice-print extraction, or emotion/sentiment inference (prohibited by design; see Section 2.3 and the EU AI Act Art. 5(1)(f) workplace emotion-recognition prohibition)
- Scoring, ranking, or evaluating your answers, or profiling you across sessions; we do not engage in automated decision-making with legal or similarly significant effects under Art. 22 (see Section 10)
4. Who We Share Data With (Sub-processors)
We engage carefully selected sub-processors. They process personal data only on our instructions and only as needed to operate the Service. The table below reflects the sub-processors used in production as of the last-updated date.
| Sub-processor | Purpose | Data processed | Location | Safeguards |
|---|---|---|---|---|
| Apple Inc. | iOS app distribution, In-App Purchase, Apple Push Notification service | Account/entitlement identifiers, subscription status, push device token | Ireland (EU) / Worldwide | EU-US Data Privacy Framework |
| OpenAI, L.L.C. | (a) Voice transcription — speech-to-text of optional voice answers (whisper-1); (b) Synthetic interviewer voice — text-to-speech (tts-1 / tts-1-hd); (c) AI generation of sample model answers and interviewer turns |
(a) the audio recording only — no identifier attached; (b)/(c) question/answer text and prompts you provide, with no account/session identifier attached at the AI-prompt layer | United States | SCCs (2021 Module 2) + EU-US Data Privacy Framework + API "no-training" / zero-retention provisions where available |
| RevenueCat, Inc. | Subscription / entitlement management | Account identifier, subscription/entitlement state | United States | EU-US Data Privacy Framework + SCCs |
How the voice path works (no identifier crosses the vendor boundary)
When you submit a voice answer, the audio is sent to OpenAI with only the audio bytes, the model name (whisper-1), a response-format flag, and an optional language hint — and nothing else. No name, email, IP, device ID, account ID, or session ID reaches the processor. After transcription, the audio is deleted; only the derived text and a provenance manifest are kept. This minimization is a verified, load-bearing control, documented in our Transfer Impact Assessment (Section 5).
We do not share data with sub-processors not listed above without prior notice and, where required, your consent.
5. International Data Transfers
Where personal data is transferred outside the EEA (to our AI processor or subscription processor in the US), we rely on:
- The EU-US Data Privacy Framework (Art. 45 adequacy) as the primary basis where the processor is self-certified
- The EU Standard Contractual Clauses (2021 Module 2) as the fallback safeguard, incorporated in the OpenAI Data Processing Addendum
- Additional technical measures: no account/session identifier attached at the AI-prompt or voice-transcription layer; vendor no-retention of audio; derived-text-only storage; encryption in transit
The voice-answer transfer to OpenAI (US) is assessed in our Transfer Impact Assessment (docs/legal/compliance/careermate-voice-tia-v1.md) and Data Protection Impact Assessment (docs/legal/compliance/careermate-voice-dpia-v1.md). Because OpenAI processes the audio in the United States, we make no claim that voice or audio is processed in the EU or stored in Europe — the transcription model has no EU endpoint.
You can request copies of the safeguards in place by emailing privacy@w4app.com.
6. How Long We Keep Data
| Data | Retention |
|---|---|
| Account data | Until account closure + 30 days |
| Interview session content (text, summaries) | Until you delete the session or your account |
| Voice answer audio | Not retained — deleted automatically after transcription |
| Derived transcript text + provenance manifest | Retained as session content until you delete it or your account |
| Subscription / billing records | 7 years (legal obligation — accounting/tax) |
| Diagnostic / crash data | 90 days |
| Support communications | 24 months |
| Minimal audit-log / event record | Retained for 5 years in pseudonymized form per EU AI Act Art. 12 + ESC-028 — the user identifier is replaced with a non-reversible pseudonym so the event trail survives without identifying you |
When you delete your account, we erase your identity, your session content, your derived transcripts, your voice-consent record, and associated on-disk artifacts; we retain only the minimal, pseudonymized audit-log record described above, plus subscription/billing records required for legal compliance (accessible only on legitimate request). See Section 7 for how to delete.
7. Your Rights (GDPR Articles 12–22)
If we process your personal data, you have the following rights:
| Right | What it means | How to exercise |
|---|---|---|
| Access (Art. 15) | Get a copy of your data | privacy@w4app.com |
| Rectification (Art. 16) | Correct inaccurate data | In-app for most fields, OR email |
| Erasure (Art. 17) | Have your data deleted | In-app account screen → Delete account, or email privacy@w4app.com |
| Restriction (Art. 18) | Limit how we process your data | |
| Portability (Art. 20) | Get your data in a portable format | |
| Objection (Art. 21) | Object to certain processing | |
| Withdraw consent (Art. 7(3)) | Withdraw consent where consent is the basis (e.g., turn off voice answers in settings) | In-app voice-consent settings, or by email |
| Lodge a complaint | Complain to a supervisory authority | Finnish DPA (Tietosuojavaltuutettu): https://tietosuoja.fi/en |
We respond within 30 days (extendable by 60 days for complex requests, with notice). We do not charge for these requests except where they are manifestly unfounded or excessive.
Account deletion fully erases your account and content (a user-initiated, in-app path required by Apple Guideline 5.1.1(v) and GDPR Art. 17), retaining only the minimal pseudonymized audit-log record in Section 6. Deletion also requires a fresh-authentication confirmation because it is irreversible.
8. Children
CareerMate is intended for job-seekers and is not directed at users under 16 years of age. (The Terms of Service set the same minimum contracting age — see Terms Section 2.) We do not knowingly collect data from children below this threshold without verifiable parental consent. If you believe a child has provided us data, contact privacy@w4app.com and we will delete it promptly.
9. Security
We implement technical and organizational measures appropriate to the risk:
- TLS / HTTPS for all network communications
- No retention of voice audio — deleted automatically after transcription; derived text only at rest
- No identifier transmitted to the transcription vendor with your audio
- No voice-print, emotion, or sentiment inference — absent by design
- Server-side key injection — API keys are never on your device
- Access controls for administrative systems
- Incident response plan with 72-hour breach notification to the supervisory authority per Art. 33
Despite our efforts, no system is fully secure. In the event of a data breach affecting your personal data, we will notify you and the supervisory authority as required by law.
10. Automated Decision-Making and Profiling
CareerMate uses AI to generate interview questions in a synthetic voice, to transcribe your spoken answers, and to produce standalone sample model answers for reference. It also surfaces objective delivery metadata (filler-word counts, pause lengths, turn timing).
We do not:
- Score, rank, rate, or characterize your answers or your "hireability"
- Compare your answer against the model answer with any diff, score, or percentage
- Route the next question based on an inferred quality estimate of your prior answers
- Build a cross-session "skill" profile or condition future content on your history
- Make decisions with legal or similarly significant effects about you under Art. 22 GDPR
CareerMate is a preparation tool, not an evaluator or assessment service. You can always disable voice features in settings and request human review of any output — email privacy@w4app.com. (Free-tier session limits are a service-availability control, not a decision producing legal or similarly significant effects.)
11. Cookies and Tracking
The mobile app does not use cookies in the web sense and does not track you.
Our marketing website (w4app.com, and any CareerMate landing page) may use:
- Essential cookies (session, language preference)
- Analytics cookies (only with consent — see cookie banner)
We do not use advertising cookies, cross-site tracking, or the IDFA. We declare our Apple Privacy Manifest accordingly.
12. Apple App Privacy Disclosures
This Privacy Policy is the canonical source. The Apple App Privacy survey (App Store Connect) must be kept in exact sync with it and with the sub-processor table in Section 4.
Apple App Privacy — declared:
- Data Linked to You: Contact info (email, for magic-link sign-in); Purchases (subscription state); User Content (interview session text/summaries)
- Data Used for App Functionality (not linked / not for tracking): Audio Data (optional voice answers → transcription only, not retained); Diagnostics (crash/performance)
- Data Used to Track You: None
The optional voice feature adds exactly one data type — Audio Data under "User Content," purpose App Functionality, not linked to identity, not used for tracking, not retained. We do not declare "Health & Fitness" or "Sensitive Info" as a purpose of collection: the incidental special-category content a user may speak is not collected for that purpose (the purpose is transcription), and GDPR Art. 9 incidental exposure is handled by the consent gate and minimization, not the Apple label.
The App Store nutrition label discloses declared-data only; this hosted Policy is the full GDPR Art. 13/14 disclosure and is the document referenced by the mandatory Privacy Policy URL field in App Store Connect.
13. Changes to This Policy
We may update this Privacy Policy. For material changes, we will notify you in-app and (if you have provided an email) by email at least 30 days before the change takes effect, and make the prior version available on request. Continued use after the effective date constitutes acceptance.
14. Change of Controller
The data controller is currently the toiminimi listed in Section 1. We plan to transfer the Service to w4app Oy in 2026 (expected Q3 2026). When this transfer occurs:
- You will be notified at least 30 days in advance via in-app message
- All your data and consent records will transfer to the new controller
- This Privacy Policy will be reissued under the new controller's name
- Your rights and the level of protection will not be reduced
The legal basis for the transfer is GDPR Art. 6(1)(b) and (f) — necessity for the contract and the legitimate interest of corporate reorganization that does not adversely affect you.
15. Contact
- General privacy questions: privacy@w4app.com
- Subject access / data export / deletion requests: privacy@w4app.com
- Complaints: Finnish Data Protection Ombudsman, https://tietosuoja.fi/en (or your local EU DPA)
Document control
- v1.0 (2026-06-19) — Published. Founder sign-off granted and licensed-attorney review waived for v1 by founder directive (ESC-044, 2026-06-19). Data controller is the natural person Ilia Vlasov operating as "w4app" (sole trader); transfer to w4app Oy planned Q3 2026 (Section 14). Hosting confirmed in AWS eu-north-1 (EEA). No third-party crash/error-monitoring sub-processor ships. Data-handling facts grounded in code-verified sources: voice INPUT → OpenAI Whisper (US, audio purged post-transcription, no identifier crosses the boundary); voice OUTPUT → OpenAI TTS, marked per EU AI Act Art-50(2); session text/summaries persisted in PostgreSQL on EC2 (eu-north-1); free tier 5 sessions/mo, Pro $9.99/mo + $89.99/yr. House style mirrored from the TravelMate / HealthMate v1 templates and the locked CareerMate voice artifacts (
careermate-voice-privacy-label-v1.md,careermate-voice-tia-v1.md,careermate-voice-dpia-v1.md).