Growth OS console

From data to decisions to actions.

A team of AI agents reads every source, tests each claim and proposes actions. The right person approves or rejects; approved actions run inside guardrails and are measured against what would have happened anyway. Every number on this page is computed from data at build time.

Agents

One governance pattern for every agent: what it reads, what it may do, how it is graded, what it costs.

Evaluation history

Synthetic data with problems planted on purpose; a separate grader checks each memo against the answer key. Failures stay in the history.