When we think about an intelligent assistant, we usually picture one thing: something that talks to us. But the real work isn’t just talking. Behind every fluent conversation, deep cognitive work has to happen: understanding intent, extracting details, and turning a vague request into a clear action. We believe these two jobs are two different layers of intelligence — and separating them, not blending them, is the key to building something that genuinely works.
Two layers, two responsibilities
The first layer is conversational intelligence: the one that faces the human, understands natural language, guides the conversation, and asks when something is unclear. This layer should be fluent, patient, and human. But it deliberately doesn’t enter the complex logic of the domain; its job is to clarify intent and gather enough context.
The second layer is background intelligence: the specialised services that do the heavy cognitive work. This layer doesn’t talk to the human; its job is deep processing — turning a request into a structure, inferring priority, and running the domain logic. This is where the real expertise lives.
Why separation matters
The easy temptation is to pour everything into one layer: a single agent that both talks and does the heavy work. But these two jobs have different natures. Conversation should be fast and fluent; deep cognitive work may be slow and heavy. If you blend the two, either the conversation slows down or the cognitive work goes shallow. Separating the layers lets each be the best version of itself.
A simple example
Imagine someone sends a message: “My order hasn’t arrived and nobody is answering.” The conversation layer takes this, understands it’s a complaint about a delay, and asks for the order number if needed. It then hands the work to the background layer, which turns it into a structured item: the issue type, the priority, and the right action. In the end, the conversation layer gives the user a clear, human reply. The user saw a simple conversation; but behind it, two layers of intelligence worked in concert.
This isn’t a chatbot plus an API
We want to be precise: this isn’t merely “a chatbot that calls a service.” The difference is that both layers are intelligent. The conversation layer isn’t just an interface; it understands intent and guides the conversation itself. The background layer isn’t just a fixed function; it reasons and decides itself. This is cognitive layering: a conversational presence alongside specialised depth, neither of which is sufficient alone.
Putting it together
We believe the future of intelligent assistants isn’t one do-everything agent, but layers of intelligence that each do their own job excellently and work in concert. The conversation layer stays with the human; the background layer brings the depth. And it’s precisely this division of labour that lets a simple message become the right action — without sacrificing either the fluency of the conversation or the depth of the expertise.