Getting started
Overview
What Converge is and how its parts fit together.
Converge is an ontology-centred data and AI operating platform. It connects to the systems an organisation already runs, turns their data into a shared model of real-world objects, and lets people, applications and AI agents work on that model through governed actions.
The ontology at the centre
Every capability in Converge reads from or writes to the same ontology:
- Data integration brings data in from databases, files, SaaS applications and streams, and lands it in versioned datasets.
- The ontology maps those datasets to object types (for example Order or Asset), the links between them, and the actions people may take on them.
- Applications such as Objects, Map, Object Analytics and custom apps show and edit ontology objects.
- Converge AI gives models and agents access to the same objects and actions, under the same permissions as the person using them.
Because everything goes through one model, a permission or marking set on data applies everywhere that data appears, including AI answers.
Governed writes
Changes to the ontology go through actions. An action validates its parameters, checks the caller's roles and submission criteria, and records the edit. Write-back to source systems happens only through approved actions or exports, never by pushing directly to a source.
AI agents follow the same path: an agent can only change data by calling an action, and write tools wait for a person to review them.
Where to go next
- Core concepts explains projects, datasets, objects, links and actions.
- Navigating the platform is a tour of the workspace.
- Platform APIs covers programmatic access.