Learning Platforms & Capability Transfer

The Craft of the Future

Where Ideas Become Systems.

The difficult idea

Help participants redesign a real entrepreneurial decision around AI without treating a single prompt or synthetic customer as evidence.

FTL role: Workshop concept, curriculum, delivery and workflow design
Domain: Universities & Institutions
Format: Workshop · Capability transfer
Credits: Developed for Social X Factor at Munich University of Applied Sciences and delivered by Maximilian Dauner through MUC.DAI.

Participants working during The Craft of the Future workshop

01
Context
The workshop moved from model foundations and bias to shared context, traceable evidence, agentic workflows and human approval.
02
Creative vision
Make AI-native work observable: context, evidence, tools, memory, evaluation and review form an intentional system.
03
Challenge & constraints
Technical ambition had to remain subordinate to real decisions, inspectable evidence and responsible use.
04
What was made
An interactive learning journey culminating in redesigned entrepreneurial workflows around real problems.
05
What was built
A practical framework distinguishing chat, fixed workflows, single agents and multi-agent systems, with clear decision gates.

Dual lens

Experience / System

The outcome and the infrastructure remain part of the same project truth.

Participants working together during The Craft of the Future workshop

What people see and feel

Work on a real decision

Participants identify what to capture, delegate, verify and retain as a human responsibility.

System

Observable workflow. Shared truth, traceable transformations, evaluation criteria and approval gates replace prompt mythology.

Human + technical orchestration

Direction remains visible at every stage.

Human responsibility

Choose the real problem
Test assumptions with real evidence
Define approval gates
Retain accountability

Technical responsibility

Capture structured context
Coordinate bounded tasks
Preserve traceability
Support prototyping

Iteration & validation

Synthetic analysis is explicitly separated from customer evidence; autonomy is bounded by access, observation and approval.

Output & deployment

Delivered as a live entrepreneurship workshop.

Reusable capability

A workshop and method organizations can use to select and govern the right level of AI workflow complexity.

Learning

Complexity should be introduced only where it creates inspectable value and preserves human control.

Presentation