The world is not only software.

Science is about to get a lot faster and more accurate. Models will be doing most, and eventually all, of the work.

Models have read nearly everything that has ever been written, but experienced close to nothing on what the world itself has said.

Experience comes from acting on a world, and eventually being graded by it. Yet, today's models have close to no experience conversing with the real world, and we believe that can be solved.

The majority of what happens on Earth is never written down. It passes through instruments as signal. We will be turning those signals into experiences, and measurements of how much frontier science they can do.

A model trained this way should be able to read an instrument, predict what it will show, and eventually operate it.

We're backed by Y Combinator, currently in the F26 batch.

We're not hiring and won't reply to applications or CVs.