> Site index: [llms.txt](https://ortelian.com/llms.txt) with all pages and descriptions.

# The platform your AI-native business runs on.

Source: https://ortelian.com/platform/
Markdown: https://ortelian.com/platform.md

Ortelian connects what your agents know about your business, so they can act in your systems and learn from each other’s work.

[Talk to us](https://ortelian.com/contact/)

[How we work](https://ortelian.com/how-we-work/)

The platform

## Give agents a world to work in.

A goal tells an agent what to achieve. To work towards it, the agent needs to understand the people, companies, events and relationships involved, and be able to act in the systems where the work happens.

Work with your agents through Slack, email or Ortelian’s chat interface. You can also access them through MCP from AI tools such as ChatGPT and Claude.

*Diagram: Your team sets goals and reviews the work through Ortelian chat, Slack, email, or AI tools such as ChatGPT and Claude via MCP. Inside Ortelian, agents use context from your company's world model and retain knowledge in it, within evaluations and action limits. Agents read and act in your systems, observe other sources to update the model, and send context to interchangeable external AI models, which return responses.*

Knowledge that carries forward

## The understanding stays with the business.

The world model brings company knowledge, market observations and experience from agent work into the same connected picture. What one agent learns can inform what another does next.

---

Knowledge about the world

Companies, people, markets and events, connected to the records, conversations and observations that describe them.

Knowledge from the work

Findings, methods and work records, retained so later work can use the experience and examine how a conclusion was reached.

Our goal is for agents to spot what needs attention and take on useful work without being told every next step.

Built for your business

## One platform, fitted to your business.

The agent software, world-model foundation and evaluations come with Ortelian. We connect your systems and develop the knowledge, working methods and workflows around your business.

Your team can give tasks, assign ongoing work and shape how it gets done. We bring the platform into the business and keep improving it with you.

Models can change. Your context carries forward.

Ortelian is built to work across AI models. We test models against your work, while the business context, knowledge and working methods remain in the platform.

Your data

## You own your data and world model.

You can export both. Ortelian hosts and operates the platform. We agree access, provider choices and data-handling requirements for your deployment.

We default to provider arrangements that do not use customer data for training. Any exception is agreed with you.

Access & accountability

## Control the access. Check the work.

Access and actions are configured for the work. We define where people must decide, keep records of execution and use evaluations to check the results.

- Workspace access

  Membership controls your team’s access. Ortelian operators manage the service.
- Agent tools

  Tools are granted to the agent. Connecting a system does not grant access by itself.
- Approval points

  We define where an action needs a person’s decision before putting the workflow into use.

*Diagram: Real examples and standards feed an evaluation that finds errors and weak decisions. Knowledge, rules or workflows are updated, the revised system is checked again, and the loop repeats after changes.*

*Illustration · evaluations check the effect of changes to the system.*

---

## Let’s work through the fit.

Bring the systems you use and the work you want AI to take on. We’ll explain how Ortelian would fit and work through your data and access requirements.

[Talk to us](https://ortelian.com/contact/)