Products
Technology should work more like people do.
We build products that begin with how people naturally think, speak, and work, while the complexity behind them stays out of the way.
Agentic Familiar
The next interface is an agent that is yours.
While most AI agents are still organized around a single company, platform, or task, we are interested in what becomes possible when the agent is organized around a person.
We call this kind of agent a familiar. It serves as your interface across applications and handles the technical complexity for you.
AI feels different when it understands how you think, knows the projects you have already done together, and can pick up where you left off. You spend less time bringing it up to speed and more time doing real work.
A familiar is not tied to one fixed job. It is built to represent a person across many kinds of work. The best work with AI is not a single interaction — it is a working relationship.
As the work changes, its capabilities can change with it. Your familiar can use different provider models, choose tools, and delegate to specialist agents without asking you to manage a digital workforce or become an AI operator.
Through that work, the familiar becomes more distinctly yours. It learns what you care about, how you make decisions, and what good work looks like to you.
Over time, a familiar becomes an extension of how you work: an intermediary between what you want to accomplish and the technology needed to carry it out. It handles the path through the technology while you keep the direction, judgment, and decisions that matter.
Co-create Workspace
A shared workspace for teams and their familiars.
Co-create is built for teams whose individual AI use is moving faster than their ability to keep shared work coherent.
Each person works with their own familiar inside a project the team shares. A conversation can become a document, plan, or working tool the team can continue shaping without losing the decisions and context behind it.
Explore Co-createLet's connect
Curious where this fits your work?
If the way we are approaching this matches something you are working on, we would like to hear about it.