Privacy Policy

Effective date: TBD (set on App Store release day) Last updated: 2026-04-19

Casemate ("the app", "we", "us") is an educational simulation tool for medical and health-professions trainees. This policy explains what we collect, why, and how you can get rid of it.

1. Who this policy applies to

Anyone who uses the app. The app is for trainees age 17 and older. It is not intended for children.

2. What we collect

Data	Source	Why we have it
Anonymous device identifier	Generated at first launch	Lets us attach your encounters and scores to a single device without requiring a login
Transcripts of your practice encounters	Your microphone, transcribed in real time	Used to score the encounter and show you the transcript in your report
Encounter metadata (case id, duration, score)	Generated by the app	History and progress tracking
Diagnostic logs (crashes, error codes)	Generated by the app	So we can fix bugs

We do not store raw audio. The microphone is used only during an active encounter, and audio is transcribed to text and then discarded.

We do not knowingly collect protected health information (PHI). All patient personas are AI-generated. You should not discuss real patients or real clinical situations inside the app.

3. How we use it

Provide the simulation (transcription, scoring, report rendering).
Show you your own history and trends.
Enforce per-device usage limits (described in §7).
Debug and improve the app.

We do not sell your data. We do not use your data to train third-party models beyond the single-turn transcription and scoring calls described in §4.

4. Processors we use

Supabase — hosts our database, Edge Functions, and authentication. Stores transcripts and scores server-side.
Vapi — voice pipeline. Transcribes your microphone input to text in real time. Vapi does not retain raw audio after the call ends, per its processor agreement.
Anthropic — LLM provider (Claude). Generates the AI patient's responses and scores the transcript. Anthropic does not use customer API traffic to train its models per its terms.
Apple — App Store distribution and in-app mechanics (anonymous identifiers, push notifications if used later).

Sharing with Vapi and Anthropic happens only during and immediately after an encounter you start, only with the data described above (real-time microphone audio for transcription; the text transcript and your entered orders for dialogue and scoring), and only after you have given explicit consent in the app's onboarding ("How your voice is processed"). Both providers are bound by agreements that prohibit retaining raw audio (Vapi) and training on customer data (Anthropic), providing protection equal to this policy. All processors are bound by their own privacy terms.

5. Retention and deletion

Transcripts and scores are retained until you delete them. An in-app delete option will arrive in v1.1; in the interim, email alichah414@gmail.com from the device you wish to wipe and we will delete server-side records tied to that anonymous id within 30 days.
Diagnostic logs are rotated out within 30 days.

6. HIPAA

We are not a covered entity and not a business associate. No real patient data should be entered into the app. If you ever accidentally enter identifying information about a real person, email alichah414@gmail.com and we will purge it.

7. Fair-use caps

We enforce per-device daily usage caps and a global daily spend cap. These are implemented in _shared/caps.ts and exist to prevent abuse, not to profile you.

8. Children

The app is for users 17 and older and is rated accordingly. We do not knowingly collect data from children under 13.

9. Your rights

Depending on where you live, you may have rights to access, correct, or delete your data. Email alichah414@gmail.com; we will respond within 30 days.

10. Changes

We will post changes here and update the "Last updated" date. Material changes will also be announced in-app at next launch.

11. Contact

alichah414@gmail.com