Skip to main content
We build Core for systems that need to learn while they operate. A stateful workflow connects a structured result to what happened next:

​
What changes through use

  • State carries forward: Later work can use constraints, relationships, and evidence formed earlier.
  • Returned outcomes can revise later behavior: Admitted observations and measured results can change which task structure carries support.
  • Results stay inspectable: Your application receives defined outputs, supporting evidence, and learned state.

​
Choose a surface

Build directly with Adapt-1, or use Rei Chat when language is part of the workflow.

​
Start building

Make an Adapt-1 call

Authenticate, choose a route, and get your first structured result.

Choose an Adapt-1 workflow

Start from the result your application needs and keep learner settings at their defaults.

Start with Rei Chat

Authenticate and make your first conversational API call.

​
Research and resources

Research direction

Read the principles behind what we are building and why.

Lab and resources

Browse our research artifacts, model releases, and supporting material.