GeneAI health platform for genetic insights

A genetic result can hold the answer and still be useless, if nobody in the room has time to read it.

GeneAI started as an idea: take genetic data and make it explainable to the people who order the tests. We built it into a working product for doctors and genetic counselors, and along the way we learned that the hardest part of health software is almost never the AI.

The idea

Before we started, there was nothing. No prototype, no old app to modernise. Just a clear problem and a belief that AI could help.

Genetic data is dense, written for specialists, and growing faster than any clinician can read. Doctors and genetic counselors needed a way to take the raw information and turn it into something they could use in a real consultation, not a stack of papers to interpret later.

So the goal was never to replace clinical judgment. It was to give the clinician an answer they could trust and act on.

The hardest part was never the AI

We knew how to build an AI backend. What we had to earn was the right to touch health data at all.

HIPAA compliance and health data safety shaped every decision: where data lives, who and what can reach it, how it is encrypted, what gets logged, and what the AI is allowed to see. Health data is not a feature you add at the end. It is the foundation the rest of the system stands on.

That is the lesson we now carry into every health product we build: treat compliance as architecture from day one, not as paperwork for later.

What we built

GeneAI is a Next.js application with an AI backend, built around three jobs a clinician actually has.

• It analyzes genetic data, turning raw information into a structured result.
• It explains findings in plain language, so the clinician can act on them.
• It summarizes research and evidence, so every answer arrives with its basis attached.

The feature we are proudest of is real-time AI voice conversation. A doctor or counselor can speak a question about a case and get a spoken answer back, without leaving the conversation to go and type a query. It turns the platform from a report you read into a colleague you can ask.

Making health data safe

Compliance is not a document you file. It is a set of boundaries inside the system.

We isolate and encrypt health data, control who and what can reach it, keep a full audit trail, and keep the AI inside a defined boundary. Every access is accountable. This part is far less exciting than the AI, and far more important. A genetics tool that leaks is not a product. It is a liability.

What changed

GeneAI is early, and it reached HIPAA certification. In health software that certification is not a badge for the marketing page. It is permission to work with real clinical data, which is the only way a tool like this ever becomes genuinely useful.

The team at Darphiz helped us simplify complex ideas into a clean, intelligent platform. Their design process was smooth and the final product perfectly reflected our vision for healthcare innovation. Patrick Roebuk, GeneAI

What we would do differently

We would treat HIPAA as architecture from day one.

We got there, and the product is certified. But compliance reaches into the data model, the storage, the APIs and the logs. Deciding those things with compliance in mind from the first commit is faster and calmer than retrofitting them later. If you are building anything that touches health data, start with the rules of the road.

Build with us

If you are building a health or AI product that has to handle sensitive data safely, from the schema up, we have done it. Talk to us.

Start a Project →