Understanding AI, and putting it to work.

We study how AI models work on the inside, and we build software that puts AI and automation to work for people. Our mission is to make technology accessible to everyone and to find new ways to use it to help build the future.

Our work in mechanistic interpretability covers ablation, removing parts of a model to see what they do; vector steering, nudging a model's internal activations to change how it behaves; and mapping where ideas sit inside a model, to see how they relate to one another.

Our software is for creators and for people working in AI, business and automation. It runs on your own device, so you control the technology and your data stays yours. Software you buy is yours to keep.