7-lesson course · about 15 hours

Run three AI agent teams on one project record, and release the handover yourself

For Consultants, project leads, managers and professionals who turn information into plans, decisions and client work

Everything for Professionals

A self-paced capstone for professionals who already brief single proactive agents. Build Discovery, Delivery Planning and Acceptance teams, carry a late scope change through them, and approve what leaves.

Try lesson 1 freeSee what you’ll learn

What you’ll be able to do

By the end, you can…

Design a Factory around one shared recordSet up three proactive AI agent teams for a bounded project, each owning one output in a shared project record. Discovery owns the checked brief and sources. Delivery Planning owns the options, plan and action register. Acceptance owns the definition of done.
Write handoff agreements that send weak work backWrite down what each team must pass on: the brief, sources, assumptions and approval status. Build a handoff checklist the receiving team uses. Work that falls short goes back to its owner with the reason recorded.
Have Dots or Claude Duet follow the approved recordSet up OpenAI Dots or Claude Duet to follow the record across team handoffs and report current actions, draft deliverables and the decisions holding up progress. Nothing gets sent, committed to or sent to anyone unless you approve it.
Carry a late scope change through every teamWhen Wednesday's out-of-scope preference arrives, log it, pause only the work it affects and hold the handover. Then put the tradeoff to the project owner: widen the scope, move the deadline or keep the requirement. The logged decision goes to all three teams.
Test the Factory against the week's surprisesRun changing inputs, blocked work, repeated inputs and a later restart from saved state. Check for duplicate tasks or messages, scope that grew without anyone deciding it, accepted requirements that were dropped, and owners, dates or approvals that were guessed rather than agreed.
Release a handover you can vouch forWrite a definition of done: every claim sourced, every dependency feasible, every commitment approved, and a page the recipient can act on without a follow-up call. Write operating rules another responsible person can inspect. Then run a full simulation from a seeded request to a reviewed handover pack.

How it works

A tutor that meets you where you are

  1. Lessons that adapt to your levelThere are seven lessons across three modules. Each one is built around the sample week: Monday's request, Tuesday's conflicting lead times, Wednesday's out-of-scope preference, Thursday's slipped test and Friday's handover. The explanations and examples adjust to what you already know.
  2. An AI tutor you can ask anythingAsk questions by typing or out loud. The tutor stands in for a live trainer. It helps you draft handoff agreements, check your record against your rules and work out why a test failed.
  3. Quizzes and flashcards for each lessonShort quizzes and flashcards are tied to each lesson, so you can check that you understand ideas like handoff status, stop conditions and the definition of done before you build on them.
  4. Progress tracking and a certificateTrack your progress through the modules at your own pace, and get a certificate of completion when you finish.

Curriculum

The curriculum

The course has seven lessons in three modules. You design the Factory, run a late change through it, and then release a handover you have reviewed. Every lesson uses the same sample supplier recommendation, so each piece you build feeds into the next.

Module 1Organize proactive AI agent teams around the outcome2 lessons

  1. 1Give each proactive AI agent team a defined output
    • Explain what a Factory of proactive AI agent teams is
    • Design teams for evidence, delivery planning and handover checking
    • Define each team's inputs, output and done criteria
    • Choose how OpenAI Dots or Claude Duet will follow the approved record
  2. 2Specify the handoff agreements
    • Write a handoff agreement between two agent teams
    • Build a handoff checklist (PDF) the receiving team uses
    • Mark where project owner approval is required
    • Define what happens when a handoff is incomplete

Module 2Carry a change through the project2 lessons

  1. 3Run the proactive AI agent team handoffs
    • Run the team handoffs from accepted request to reviewed delivery plan
    • Test each handoff against its checklist
    • Review the current actions, draft deliverables and decisions holding up progress returned by the tool
    • Introduce a late change and record how the teams respond
  2. 4Resolve conflicting requirements
    • Identify conflicts between requirements or between a change and an accepted requirement
    • Trace which deliverables, actions and dates a change affects
    • Prepare options for the project owner with evidence, judgment and approval separated
    • Test that the approved revision is applied consistently across teams

Module 3Deliver a system you can direct3 lessons

  1. 5Refine the proactive AI agent team operating rules
    • Write operating rules covering permissions, approvals, data boundaries and update cycles
    • Define stop conditions where agents pause and escalate
    • Revise rules based on findings from earlier tests
    • Explain the Factory's limits in plain language
  2. 6Define the handover decision
    • Define what counts as an accepted handover for the sample project
    • Assemble a handover package that another person can inspect
    • Identify who approves the handover and what they must review
    • List open risks and decisions that travel with the handover
  3. 7Run the complete project simulation
    • Run the Factory from initial request to accepted handover
    • Test changing inputs, blocked work and a later restart in one simulation
    • List what a manager must review after each test
    • Present the results, limits and approval points for inspection

Who it’s for

Made for you if…

Made for: Consultants, project leads, managers and professionals who turn information into plans, decisions and client work

Consultants and project leads

You already brief single agents. Now you need them to stay consistent with each other, so a change your evidence agent picks up also reaches your plan and your draft before anything goes to a client.

Managers who own the release

You are accountable for whatever leaves under your name. You want a written record of what was accepted, on what evidence and by whom, and approval points that another person can inspect.

Analysts and operations or procurement professionals

You turn unclear requests into plans and recommendations, often alone or in a small team. You want agent teams to do the middle of the work while you keep scope, commitments and stakeholder communication.

Questions

Good to know

What do I need to know before starting?

You should be able to turn a one-line request into a brief covering audience, purpose, deliverables, exclusions and acceptance criteria. You should also have briefed at least one proactive agent with a trigger, in-bounds sources and a rule for what goes back to a person. That is Days 1–7 of the Professionals track or equivalent experience. This course doesn't teach those skills from scratch.

Which tools do I need?

You need access to OpenAI Dots or Claude Duet. The course treats them as equals, so pick one. You also need a workspace for the shared project record, such as Notion AI or Microsoft 365 Copilot.

Can I use a project from my own work?

Yes, as long as you have permission to use the material and keep scope, commitments and review responsibility clear. Otherwise you work on the course's sample project: a supplier recommendation for an operations director.

Will the agents contact anyone or make commitments?

No. The course sets them up to sort, draft, flag and report. You, as project owner, approve scope, commitments and stakeholder communication. The course doesn't involve live client engagements or sending real stakeholder communications, and client-confidential or employee data stays out of connections that haven't been approved.

Do I need to write code?

No. The course doesn't cover coding custom multi-agent frameworks or API integrations. You design teams, handoffs and rules in plain language and run them in the tools listed above.

How long does it take?

About 15 hours, self-paced.

What does the course leave out?

It doesn't cover comparing tool pricing or licensing, formal project-management methods or certifications, or organization-wide AI strategy and change programs.

Build a Factory of agent teams and release a handover you approved

$397

One payment · lifetime access

  • 7 self-paced lessons in 3 modules, about 15 hours
  • A sample supplier-recommendation project to work on, or your own if you have permission
  • Hands-on setup with OpenAI Dots or Claude Duet, your choice
  • Tests for late scope changes, blocked work, repeated inputs and a later restart
  • An AI tutor plus quizzes and flashcards for each lesson
  • Progress tracking and a certificate of completion

Secure checkout by Stripe. Have a promotion code? Enter it at checkout.

Refunds: Full refund if you ask before you open your first lesson. Once you’ve started, purchases aren’t refundable, except where the law gives you that right.

Already bought it? Sign in