FAQs
No. Depending on the solution, the AI reads and understands incoming information, reasons over your business context to decide the right action, extracts or generates the specific content needed, and — within the boundaries you set — acts inside your systems. A chat interface may be part of that (for knowledge retrieval, for example), but the underlying work is understanding, deciding and acting, not just conversing.
Yes. The listed solutions cover the most common patterns we see, and Custom AI Agent System and Custom AI Application exist specifically for work that doesn't fit those patterns — a multi-step agent, or a full application. When even those don't fit cleanly, we design something around the specific way your business works.
Yes. We connect to the systems you already use through APIs, webhooks or native connectors rather than asking you to migrate to a new platform.
In most cases, yes, provided the application exposes an API or another integration point. We assess this during the discovery phase.
No. We work with your existing CRM and keep it up to date rather than replacing it.
Yes. We can configure solutions around a model or provider you already have an agreement with, where technically compatible.
Yes, open-source and private models are supported where they meet the performance and deployment requirements of the workflow.
Yes, private and hybrid deployment options are available depending on your requirements.
Yes, this is a common requirement and one of the factors that shapes which technology and deployment model we recommend.
It depends on the scope and complexity of the workflow. A single, well-defined workflow is typically faster to deliver than a multi-system programme — we scope timelines during discovery.
Yes. Approval points are configurable, and we recommend keeping a person in the loop for sensitive or high-value decisions.
We agree on the metrics that matter for your process — turnaround time, error rate, volume handled, manual effort reduced — before implementation begins.
We monitor outcomes, identify where the workflow can be improved, and adjust rules and scope as your needs evolve.