Data for biological superintelligence.

We are building a lab to collect the data models need to learn biology. We think this is how every disease gets cured.

Why we exist

§1 · Observation

Mathematics fell to models once a machine could check their work.

A proof is either right or wrong, and a computer can tell you which. Nothing can tell you that about a cell. If you want to know what a drug does to a cancer cell, you have to put it on one and watch.

§2 · Thesis

Smarter models will run more experiments.

A better model has more ideas worth testing, and it gets more out of each result. So the smarter models become, the more experiments they will want. We expect the pace of biology to be set by how fast ideas can be tested.

§3 · Value

An experiment will be worth what a model can learn from it.

Until now an experiment was run to answer one person’s question, and most of what else happened in the dish was never recorded. A model can use all of it. Once models do the reasoning, the experiments worth the most will be the ones they can learn the most from.

§4 · Petrichor

We are building for that from day one.

Petrichor is a lab where the reader is a model. We record what was done to the cells, how they behaved in the hours that followed, and what changed inside them. We are starting with cancer.

§5 · Scale

Curing every disease will need data from every disease.

Each disease is a different way for a cell to go wrong. Models will need to see the cells that carry each one, with healthy cells beside them. That data does not exist yet, so we are going to collect it.

§6 · The goal

Biological superintelligence will be built out of experiments.

This is the most important data we can think of collecting. We started Petrichor to collect it, so that humanity reaches biological superintelligence sooner and, eventually, cures every disease.

Contact us

Signed Y Combinator · Fall 2026

Wilhelm Hedenskog
CEO
Elias Reinfeldt
Co-founder
Jesper Johansson
Co-founder