thirdmind
Contact
thirdmind · Workshop
← Services · Department · Process work · on site or remotely

AI workshop for concrete processes.

An AI workshop is useful when a process is already on the table: a support process, back office workflow, finance check, sales task, or another specific work area.

We examine that process with the department, IT, and decision-makers, then clarify what implementation would require: task, data, systems, edge cases, rights, approvals, and MVP scope.

01 Who the workshop is suitable for

When the process is already on the table.

An AI workshop is suitable if you are no longer looking broadly for AI potential, but want to examine a specific process in more detail.

We look at recurring work in support, finance, sales, or operations and bring the department, IT, and decision-makers together. The goal is a sound basis for an MVP, pilot, or internal implementation.

02 Formats

Process clarification when the process is known. Implementation basis when an MVP needs to become tangible.

Format 01 Process clarification
Compact

A process map becomes a testable scope.

The compact format is suitable when the process is known, but the task, data, roles and boundaries are not yet clearly described.

  • 01 Clarify the initial situation and goal
  • 02 Make process steps visible
  • 03 Collect inputs, outputs, and edge cases
  • 04 Name data sources and target systems
  • 05 Record open questions for the department and IT
Format 02 Implementation basis
In-depth

Sharpen the process so that an MVP becomes possible.

The in-depth format moves closer to implementation. We clarify data, rights, escalations, quality criteria, and a realistic MVP scope.

  • 01 Describe specific tasks for the digital employee
  • 02 Classify system access and data sources
  • 03 Clarify roles, rights and approvals
  • 04 Collect test cases and quality criteria
  • 05 Outline MVP scope and next implementation steps
03 What happens in the workshop

A good AI workshop makes a process ready for implementation.

  • 01

    Where does recurring manual effort arise?

  • 02

    Which tasks take time without requiring new decisions?

  • 03

    What information needs to be gathered today?

  • 04

    Which systems are involved?

  • 05

    What steps could a digital employee prepare?

  • 06

    What cases do people need to review or approve?

The workshop is not meant to produce as many AI ideas as possible. It clarifies one specific process enough to make a decision: build it, prepare it internally, or stop it for now.

04 Possible workshop focuses

Four directions, depending on the process.

01 / Sharpen process

Sharpen process

For teams that know which process should be considered, but have not yet formulated a clear AI task, boundaries or success criteria.

02 / Bring in the department

Bring in the department

For processes where domain knowledge, edge cases, and daily reality matter as much as the technical solution.

03 / Clarify data & systems

Clarify data & systems

For processes involving CRM, ERP, tickets, emails, documents, tables, or internal databases.

04 / Prepare MVP

Prepare MVP

For companies that want to know after the workshop what exactly needs to be built, tested, and approved.

05 Experience

Workshop experience from projects where AI touches real work.

The workshops come from thirdmind's implementation perspective: digital employees need tasks, data, rights, system context, and clear handovers.

This background shapes the work in the room. The goal is not to make AI seem impressive. It is to understand a process well enough for a viable next step.

06 Result

What is on the table at the end.

Depending on the format, a workshop goes to different depth. The central question stays the same: is this process ready for the next step?

  • 01

    sharpened process map

  • 02

    concrete task for the digital employee

  • 03

    relevant data sources and target systems

  • 04

    edge cases, risks, and approval points

  • 05

    first test cases and quality criteria

  • 06

    MVP scope or implementation option

  • 07

    open questions for IT, data protection, or specialist departments

A workshop is not a substitute for implementation. But it can prevent an AI project from starting with an unclear task, incorrect scope or missing approvals.

07 What we don't do

Not a long list of ideas. No promise that this process should be automated.

We do not run workshops that end with only a long list of ideas. We also do not assume that every process should be automated.

Sometimes the best outcome is to clarify the process internally first. Sometimes a small MVP makes more sense than a large rollout.

08 Frequently asked questions

What is usually asked before a workshop.

How long does an AI workshop last?

Workshop length depends on scope. Compact formats focus on process clarification; deeper formats cover data, edge cases, rights, approvals, and MVP scope.

Who should participate?

It makes sense to involve the relevant department, the process owner, IT or data protection, and someone who can decide on the next implementation step.

Is the workshop about tools?

Only at the edge. The focus is a specific process: task, data, systems, edge cases, rights, approvals, and quality criteria. Tools come later.

Can the workshop take place remotely?

Yes. Workshops can take place remotely or on site.

Is a workshop already a project start?

Not automatically. A workshop creates a better basis for implementation. If the process, data, rights, and MVP scope are clear enough, a pilot can emerge.

What distinguishes the workshop from the AI Compass?

The AI Compass evaluates several possible use cases and provides a decision basis. The workshop goes deeper into one specific process and clarifies what an implementation or MVP would require.

Request a workshop

If you want to move a concrete process toward AI, the workshop is the right place.

Request an AI workshop

Yuno asks four short questions about the process, subject area, data, and goal. After that, it is clearer whether an AI workshop or the AI Compass makes sense first.