Getting started

This is a doorway, not a manual. It explains what you have, what it does, and where to look next.

What you have

You have an AI employee. Not a chatbot bolted onto your website, and not a tool your team has to learn to operate. It is a digital team member with a defined role, working in a workspace that belongs to you alone.

We set it up, we run it, and we are accountable for how it behaves. You do not manage infrastructure, you do not maintain anything, and you do not need anyone technical on your side to keep it working.

What you do is watch it work, and step in when you want to.

What it does

Your AI employee handles the role you scoped with us. The scope is agreed with you before it ever speaks to a customer, and it is narrow on purpose.

Two things follow from that, and they matter more than any feature:

It has a boundary. There are things it is not there to do, and that is by design rather than by omission. Conversations that go beyond the agreed scope are meant to reach a person instead, and the handoff queue in your console is where you see that happen.

You can see what it did. Every conversation it has is readable. You are never in a position of trusting a summary of your own customer conversations.

What the console gives you

Three things, and this page describes exactly these rather than a longer list of what is planned:

• Read every conversation. Filter them, open one, and read the exchange turn by turn. See Watching your agent.
• Take over when you want to. Step into a live conversation and reply as yourself, with your name attached. See Taking over a conversation.
• See your plan and what has been used. The billing screen is read-only and tells you where you stand.

Getting in

Signing in has its own walkthrough: Signing in. The short version is that we send you a set-password link, you choose a password once, and from then on you sign in at the console with your email and that password.

If you do not have access yet, that is a conversation with us rather than a form to find. Your account manager arranges it with you directly.

What to expect in the first weeks

The first period is about calibration rather than volume. Your agent starts with a narrow, well-understood slice of the role, and the boundary widens as it earns it. That is deliberate: an agent given the whole role on day one has no evidence behind any of its decisions.

Expect to read conversations early on. Not because something is wrong, but because reading them is how you find out whether the boundary is in the right place. If you see it handling something you would rather it escalated, or escalating something it could handle, tell us. That feedback is what moves the line.

Expect gaps. There are things the product does not do yet, and we would rather you read them on one page than discover them one at a time. See What is not here yet.

When something looks wrong

Tell us, and point at the conversation. Every claim on these pages describes something you can go and look at yourself, so if a page and the product disagree, the page is what gets fixed.