Florentin Purcea

LiveConsumer health product, live

A lab-results explainer, GDPR art. 9 by design

Manual work

  1. Issue detected

    People google each value on their lab report, one by one, in fear — and health data is the most sensitive category there is.

  2. Verification

    Verified end to end with real reports, including the trend across reports of the same person; every value cites its page and row on the source PDF.

  3. Root cause

    A search engine cannot read the interval printed on that specific report, and a model left alone will guess numbers. Trust needs the number to come from the paper, not from the model.

  4. Fix

    A deterministic parser reads every value and interval from the PDF; the model only writes the words; explicit health-data consent before anything; one-click export and erasure; one cookie, no third parties. It never diagnoses.

  5. Measured Delivered

    Live · zero analytics, zero third-party requests

    Window: productionCaveat: Usage is not measured — by design. No analytics on health data.

What was built

The system, in buyer words.

Upload the PDF; a deterministic parser reads every value and the interval printed on that report; the model only writes the words; every value cites its page and row; explicit health-data consent before anything; one-click export and erasure; one cookie, no third parties. Never diagnoses.

Built on
Next.jsCloudflare WorkersD1R2Claude
Analyse my businessWhatsApp