Press kit

Descriptions, facts and media for anyone writing about VitaCrypt or its Lab Agent. Everything here may be used for coverage. Every screen shows a fictional person with synthetic data.

01 · Descriptions

Copy you can use

One line

VitaCrypt reads your DNA, labs, wearables and local air as one health profile, and its Lab Agent follows up on every new blood test.

About 50 words

VitaCrypt is personal health intelligence: it joins DNA, blood tests, wearables, microbiome, check-ins and local air quality into one profile and checks every link against published research. Its Lab Agent reads each new lab, plans habits, repeat tests and clinician questions, and at the next lab checks whether that plan worked.

About 120 words

VitaCrypt is an early-stage personal health intelligence MVP built by solo founder Paul Burg. It reads DNA, blood tests, wearables, microbiome reports, questionnaires and local air quality as one profile. Each link it finds is checked against PubMed and Europe PMC and labelled Cited or Hypothesis. The Lab Agent follows up on every new blood test in seven fixed steps: it compares the lab with earlier ones, links changes to the person's DNA, searches the research and shows every query, then plans habits, repeat tests and questions for a clinician. At the next lab it checks its own plan against the habit log and keeps, swaps or retires each habit. It gives no diagnoses and never sets a dose. A guided demo with synthetic data needs no sign-up.

02 · Facts

At a glance

Stage: Early-stage MVP. Pre-revenue, self-funded. Real lab uploads stay closed until the founder beta.
Founder: Paul Burg, solo founder. Designed and built the platform, the Lab Agent and the live demo.
Live demo: app.vitacrypt.xyz/demo: a fictional person, Alex, with synthetic data. No sign-up; each visitor gets a private copy, deleted after 7 days.
Data sources: DNA, blood tests, wearables, microbiome, questionnaires, local air quality and climate.
Research: PubMed and Europe PMC. Each insight is labelled Cited (with PMIDs) or Hypothesis.
Agent run: Seven fixed steps per lab. A language model judges papers and words the result; fixed rules decide what the agent does.
Actions: Retest reminders, labs nearby, a repeat lab order (US sandbox), a private link for a clinician. Each one needs the person's tap. In the demo, orders and lab listings are simulated.
Privacy: Only coded facts reach the AI model, never a name, file or exact values. Health data is encrypted at rest under a per-user key held by our server; deleting data destroys that key.
Limits: Not medical advice and not a medical device. No diagnoses. Supplements appear only as official reference values with their source, never as a dose.