Vithos

Vithos is built by three people: two technical co-founders who write the product, and me as founding product designer. I designed the mobile app, the website, and the flows through both, and I ran critique and testing on the mobile side throughout.

Families end up managing health for more than one person, a parent's checkup alongside your own. The record of it is a folder of PDFs from whichever lab was nearest at the time.

A single report gives you a value and a reference range. The question people actually have is whether it is better or worse than last time, and last time is in another file, from another lab, formatted differently.

Fixing that requires every report, for everyone in the family. It is a high-friction ask: handing over a parent's blood work to an app found last week.

So the product needs the records to be useful, and has to be trusted before it is given them. Every decision below sits on one side of that trade or the other.

A report goes in as a file or a photograph taken on a phone. Most arrive as photographs, because the report is handed over on paper. The biomarkers are pulled out, attached to the right person in the family, and placed on a timeline with everything else that person has ever had measured. From there the history can be read, or asked questions against.

Graphs before numbers. A full panel is dozens of values. Set as a table, the reader compares every row themselves and gives up around the eighth. A chart does the comparison: the shape reads as better or worse before any figure is. The numbers stay one step away for anyone who wants them.

Rejected The panel as a table, which is what every lab report already is.

Zoom

A phone screen from the app. A back control and a bookmark control sit above a heading reading Vitamin D, then the current value, 18.2 nanograms per millilitre, then the reference range as Min 12 and Max 24. Below that a section headed Trends offers six ranges, 3 months, 6 months, 1 year, 3 years, 5 years and Full, with 6 months selected. The chart beneath is six monthly bars from January to June, four of them green, April amber and June red, so the direction of travel and the months out of range read before any figure does. A heading reading About Marker follows.

Say what happens to the data where the data is asked for. “Your health data is not our product” does nothing in a privacy policy, which is read long before or long after the upload. It belongs beside the upload control.

Rejected A link to the privacy policy.

The account says when it has been opened. A family account holds several people's medical records, and the person most likely to open it after you is someone in the family. Access notification is therefore a product feature, and several decisions in the sign-in flow follow from that.

Rejected Treating access as a security setting.

Zoom

An email from Vithos. It opens Hi Ashley, then a heading reading New sign-in to your Vithos account. The body reads: we noticed a new sign-in to your Vithos account, and because your account holds sensitive health information, we want to make sure it was you. A panel lists the device, a Nothing phone 2a; the location, Mumbai, India; an IP address; and the time, June 29 2026 at 10:30 PM. Below it: if this was you, no further action is needed, and in bold, if you did not sign in, please secure your account immediately by resetting your password. A Reset your password button follows, then a sign-off from the Vithos Security Team.

Interface quality is part of the trust argument. This is the one I had to defend most, because it sounds like polish for its own sake. In consumer health, software that looks careless reads as careless with the data behind it. That is where the effort went.

Rejected Shipping rough and polishing after beta.

Several rounds across every flow, weighted to mobile, where the product is actually used. The realistic moment is someone photographing a report in a clinic waiting room. I ran the critique and testing on that side myself.

The landing page is recent and had a review of its own, aimed at whether the message lands. For a product that opens by asking for medical records, it carries the trust argument before anyone has seen a screen of the app.

Zoom

A section of the vithos.in landing page. An eyebrow reads Our Promise, above a heading reading: your health data is not our product. Beneath it: we built Vithos because we wanted this for our own families, which means treating your data with the same care we would want for ours. A row of four chips reads DPDP Act, compliant, in progress; data secure, AES-256 at rest and in transit; India data residency; and right to erasure. Three cards follow. AES-256 Encryption: all health data is encrypted at rest and in transit, the same standard used by banks, and nobody reads your reports except you. Data stays in India: data is stored on servers in Mumbai and never leaves Indian jurisdiction. Zero data selling: the data is not sold to insurers, advertisers or pharma companies, not individually and not anonymised. A final panel, Our data promise to you, states that everything can be exported and deleted at any time and is gone from the servers within thirty days, with no dark patterns and no hidden retention, and carries a Delete your account link.

In beta. The mobile app has 14+ testers and has not been released yet; the web app has about ten. Extraction has been near-perfect so far, measured by comparing the extracted biomarkers against the source reports.

• 14+ mobile beta testers
• ~10 on web, both in beta
• Extraction checked against source

vithos.in

These are beta figures from a product whose app has not shipped. The mobile release is next.