What are enterprise AI agents?
Enterprise AI agents are software that plans and completes multi-step tasks using your company tools and data, within rules you set. Because they take actions, they need permissions, limits and a record of what they did.
How is Orbyqa different from a chatbot?
A chatbot answers questions. An Orbyqa agent can read documents, check records in other systems, apply your rules and pass work along, and it pauses for human approval before sensitive steps.
Can Orbyqa run on-premise or self-hosted?
Yes. Orbyqa can run on your own servers, in your private cloud account, or in an air-gapped environment, so prompts, data and logs stay where you choose.
Which AI models can I use?
Orbyqa is model-flexible. You can use hosted models or models you run yourself, and choose a different one for each agent.
How do you keep agents from taking risky actions?
Each agent has scoped permissions, allow lists for tools and actions, spending caps and stop conditions. Sensitive steps can require human approval.
What can agents connect to?
Agents connect to internal APIs, databases, ticketing systems, email and file stores using scoped credentials that you control.
Can we see what an agent did?
Yes. Every prompt, tool call, approval and change is recorded in an audit log that you can export to your own monitoring tools.
How do we get started?
Email us one workflow you want to automate. We will walk through how it would run in Orbyqa and what it needs to connect to.