A model of your metabolism you can run forward, question, and check against your own data.

Built for people who want to know what is happening in their bodies.

Wearables, continuous glucose monitors and home lab panels now give individuals more physiological data than most clinics collected a decade ago. The analysis behind them is statistical. It reports what happened and cannot say why, or what a different choice would have done.

SomaTapestry is designed for the people who take that data seriously: biohackers and quantified-self practitioners who track their supplement and medication stacks, people on GLP-1 medication who want to understand their trajectory, athletes and longevity-focused individuals who run their own experiments, and research groups who need a mechanistic model to test hypotheses against.

A wellness and research tool, by design.

SomaTapestry is positioned deliberately. It supports self-directed people and research teams who want mechanism, prediction and explanation. It works alongside clinical care rather than replacing it: medication is modelled around a regimen set by the prescriber, and the platform's role is to show what that regimen does in the body and how lifestyle choices interact with it.

That positioning sets the roadmap. Consumer and research use come first, with wearable and lab integrations expanding toward the instruments research groups already use. The platform page describes what is available today and what is planned.

Three layers, one organism.

• A mechanistic model

  A graph-based representation of human physiology that simulates glucose, insulin, hormones, lipids, body composition, drugs and lifestyle inputs forward in time, with uncertainty bands on every output.

• A cited knowledge base

  Every mechanism and parameter is tied to a primary source, a decision record, and the validation cases it must keep passing. The model can show where each number came from.

• A self-expanding process

  AI agents read the literature, propose additions, and calibrate parameters. Proposals enter the model only after automated validation and human review, so the model grows without losing its provenance.

What it does.

• Simulate your stack

  Build supplements and medications into a personal model. See interactions, enzyme load, when each substance peaks and clears, and whether a change of timing or dose spacing would help.

• Forecast meals

  Log a meal by photo, text or catalogue and see the predicted glucose response before it happens, anchored to the latest sensor reading.

• Run what-if experiments

  Fork the model at any moment, change one thing, and compare the two trajectories side by side. Each result shows its confidence and can be traced back to mechanism.

• Calibrate from your data

  Glucose, heart-rate variability, sleep and body composition from connected devices anchor the model to measured reality and refine it after every meal.

• Understand GLP-1 trajectories

  See projected weight and body-composition change, the fat-versus-lean split, and what protects lean mass during treatment and after it ends.

• Support research

  Use the same engine to test hypotheses against cohort data, with full provenance for every parameter and every run. More for partners and research groups.

Apps for every day, an engine for research.

SomaTapestry is delivered as native Android and iOS apps and a web application, all backed by the same engine and the same account. The apps cover the daily routine: logging, live forecasts, stack management and insights. The web application adds longer-horizon analysis and export. Partners and research groups access the engine directly through an API. The platform page has an overview of each.

Join the waiting list.

Access is opening in stages. Leave your email and we will let you know when your place is ready.