How we work

Map it. Build it. Run it.

Your first agent is live in about two weeks, and it takes a few hours of your time in total. After that, we keep it running.

Example

Chasing late invoices

By hand3 hours a week

With an agentAbout 10 minutes

The agent sends polite reminders. You approve the ones to your biggest clients.

  1. 01

    Map

    We sit with your team, look at how work actually gets done, and pick the one process where an agent will make the biggest difference.

    • Watch the process as it really happens, exceptions included
    • List the tools and information it touches
    • Agree what a good result looks like before we build

    Your part: An hour or two with the people who do the work today.

  2. 02

    Build

    We build the agent, connect it to your tools and test it on your real data, with approval steps wherever a person should have the final say.

    • Connect the agent to the tools you already use
    • Test it on your real work, not made-up examples
    • Add approval steps wherever a person should decide

    Your part: A short review with us before it goes live.

  3. 03

    Run

    We keep it running, fix it when your tools change and send you a regular report of what it did and the time it saved.

    • Fix and update it when your tools or process change
    • Send you a regular report of what it did and the time it saved
    • Check in with you, and help you hand over the next job when you're ready

    Your part: Read the report, and tell us what's next.

You stay in control

It asks first

Anything sensitive, like sending, paying or deleting, waits for a person to approve it.

Every step is logged

You can see what the agent did, when and why, whenever you want to check.

Only the access it needs

Each agent can reach the tools and information its job needs, and nothing else.

Start with one process.

Tell us where the busywork piles up. We'll map it with you and have an agent live in about two weeks.

Get in touch