Skip to main content
Ongoing engagementAnthropic Partner NetworkOfficial OpenAI Partner

Your AI department.Without hiring one.

A standing team that trains your leaders, builds what they ask for, and keeps your stack current as Claude and ChatGPT change every month. Same Slack, same standups, same deadlines as the rest of your company.

A practical 30 minutes: what fits, what to do first, and who does what.

Monthly
Engagement, not a project
Leadership
Cohort to start
1:1
Founder access
01

Why an ongoing team beats a one-off project

AI adoption does not fail at the demo. It fails in month three, when the person who championed it gets busy, the models change, and nobody owns the next step.

Most of our clients start with a single workshop and find the same thing: the team leaves excited, tries it for two weeks, and drifts back to the old way. The capability was real. What was missing was someone whose job it is to keep going.

That is what this engagement is. We start with a small leadership cohort, because the people who set direction are the ones whose habits spread. We build the first agents and automations with them, on their real work. Then we stay: new releases get tested before they reach your team, new requests get scoped and shipped, and the next group gets trained by people who already know your business.

We deploy both Claude and ChatGPT, so the advice is about what fits the job rather than what we resell.

02

What is included

One team, one relationship. The mix shifts month to month as you move from learning to building.
01

Leadership cohort

A small group of the people who set direction, trained over several sessions on their own documents and decisions. They leave able to use Claude and ChatGPT daily, and able to judge what the rest of the company should do with them.

02

Agents and automations, shipped

The things the cohort asks for become builds: skills, MCP connectors to your systems, agents that run a workflow end to end. Scoped in writing before we start, so everyone agrees on done.

03

Release watch

Claude and ChatGPT ship something that matters most months. We test it against your workflows and tell you what to adopt, what to ignore, and what just broke.

04

A shared channel and a direct line

We work in your Slack or Teams. Your leadership has a direct line to the founders for the questions that do not fit a ticket.

05

Governance that keeps pace

Access, data handling and usage policy set up once and revisited as adoption spreads, so security signs off instead of slowing down.

06

Monthly read-out

What shipped, what people are actually using, where the next hours should go. Written down, so the value is visible to people who were not in the room.

Want to see what the first month would look like?

Leave an email and we will send back a first-month plan for a team your size: who to train first, and the first two things we would build.

No sequence, no spam. Or book a call directly.

03

Who this is for

It fits best when the company has decided AI matters and needs someone to own making it real.

Companies from a few dozen people to several thousand, rolling out Claude or ChatGPT Enterprise, who want adoption past the first month
Leadership teams who want to be fluent themselves before they ask the company to change
Operators with a backlog of "could AI do this?" requests and no one to build them
Teams already using AI in pockets who need one standard, one set of tools, one owner
04

How it runs

Train first, then build. After training, your team builds most of what it needs itself. We build what they cannot or will not.
01Week 1

Map

Short interviews with leadership and a look at the tools you already pay for. We come back with where AI will save the most hours, and who should learn first.

02Weeks 2–8

Train

The leadership cohort runs on real work. Every session ends with something each person uses the next morning.

03From week 3

Build

The first requests become scoped builds: a skill, a connector, an agent. Each one ships with a short guide and a named owner on your side.

04Ongoing

Expand

The next teams get trained, the next builds get queued, and the monthly read-out decides where the hours go.

05

Who does what

We are specific about this up front, because a program that is done with you feels very different from one done for you.
We do
  • Design and run every training session
  • Build, deploy and maintain the agents, skills and connectors
  • Test new model releases before your team sees them
  • Write the monthly read-out
You bring
  • Name an internal owner we meet with weekly
  • Give the cohort the time to attend and practice
  • Bring real work to the sessions, not hypotheticals
  • Grant access to the systems we connect to
06

Frequently Asked QuestionsWhat teams ask us before they start.

How is this different from hiring an AI consultant for a project?
A project ends when the deliverable ships. This engagement is built for what comes after: adoption, the next request, the next model release. Clients use it as their standing AI department, with specific builds scoped on top as they come up.
Why start with leadership instead of the whole company?
Because leadership habits spread and company-wide mandates mostly do not. A small cohort of the people who set direction becomes fluent first, then decides with us which teams go next and what they need. It is faster and cheaper than training everyone at once and hoping it sticks.
Do you work with Claude, ChatGPT, or both?
Both. We are an Anthropic Partner Network member and an official OpenAI partner. Most clients end up using each for different jobs, and we train on which to reach for when.
What does it cost?
It depends on team size and how much building is in scope. We scope it on a 30-minute call and send a written proposal. Builds beyond the monthly scope are quoted separately, before any work starts.
Can we stop if it is not working?
Yes. Everything we build lives in your accounts, so if you stop, you keep all of it. We agree the term and exit terms in writing before we start.

Talk to us about an ongoing AI team

Thirty minutes. Tell us where AI is today in your company and where you want it in six months. We will tell you honestly whether an ongoing engagement is the right shape, or whether a single workshop will do.