AI Workshop in NRW: Why Germany's industrial heartland must act now

So, are you excited about GPT-5.3 Codex yet? Oh, you don't even have a plan for using AI in your business in a structured way?

Don't panic. It's not about using the biggest, fastest, best model. It's about solving real problems in your business with AI. And for that, you don't need a consultant from Frankfurt who doesn't know your industry. You need someone who understands SMEs in NRW. Someone who knows what makes a mechanical engineering firm in Remscheid tick, what a supplier in Solingen needs, and why sales in Wuppertal faces different challenges than a start-up in Berlin.

NRW is industrial territory. AI is an industrial topic.

North Rhine-Westphalia is Europe's largest industrial region. Mechanical engineering, automotive, chemicals, steel, logistics. This is where things get produced, manufactured, delivered. And that's exactly where the potential for AI lies. Not in trendy apps or chatbots, but in quality control, predictive maintenance, quotation preparation, production planning.

The reality looks different, though. Over 50% of companies are stuck in the experimentation phase. CoPilot licences got rolled out, ChatGPT Pro subscriptions handed out. Spray and pray. Three months later the licences get cancelled again because nobody saw any real value.

The problem isn't the technology. The models are there. The tools are there. What's missing is the right approach, and that doesn't start with a software licence. It starts with a workshop.

What KI.NRW does, and what it doesn't

NRW has its own competence platform for artificial intelligence, KI.NRW. That's a good thing in principle. There are funding programmes, networks, events, information material. If you want an overview, that's where you'll find your starting points.

But KI.NRW is institutional. The processes are slow, the offerings broad, the waiting times long. For a company that wants to get a concrete use case off the ground right now, that's often not enough. You get orientation, not a working prototype. You get networking contacts, not a concrete plan for your production line.

Same with the IHK Bergisches Land. Good events, informative talks, useful initial advice. But if you want to know on Monday whether AI can improve your quality control, you need someone who dives into the process with you and your team. Not in three months. Now.

Why AI workshops are different from other workshops

AI workshops have nothing to do with a retro, a strategy meeting, or a C-level retreat. I've run more than 20 AI workshops over the past few months, and learned two things along the way.

AI workshops only make sense as a package. In SMEs, the level of AI experience varies wildly across a single team. While one person could easily pass as a prompt engineer, others struggle to tell Miro apart from PowerPoint. A single workshop can't close that gap. It takes a structured process.

AI workshops are more intense than other workshops. New subject matter, wildly different knowledge levels, and the need to land on concrete results by the end, that combination makes AI workshops demanding. For the participants, and for the facilitator.

Most AI workshops out there kick off with Ghibli-style image generation, action figures, or Veo3 videos. Fun gimmicks. Impressive demos. But does any of that move your machinery forward? Does it help your sales team write better proposals? Does it improve quality control on your shop floor? No.

A good AI workshop starts with the business problem. Not the technology.

The AI Design Sprint: from problem to prototype

The AI Design Sprint is a structured process across four workshops:

1. Opportunity Mapping: C-level executives and department heads identify real pain points. What cost you 90 minutes of wasted time in the last two weeks? What's repetitive? What's frustrating? This session typically produces 10 to 15 concrete pain points. We prioritise three of them.

2. Framing Session: Analyse the most important pain point at department level. Not the biggest one, the most solvable one. What steps are involved? Where do errors, delays, and frustration come from?

3. AI Concept Development: The team develops the AI solution itself. Input, output, data sources, interfaces. This is also where I bury ideas that are too expensive, too slow, or technically not feasible. Sounds harsh, but it saves a lot of money and a lot of awkward meetings with the board.

4. Prototyping: IT check, first prototype with real data. Not a polished demo, a working PoC that proves whether the idea holds up.

After 3-4 days of workshops and a follow-up PoC phase, you have a working prototype, a team that can apply the methodology on its own, and a prioritised list of further use cases.

NRW has a funding ecosystem most people don't know about

"AI is too expensive for us!" I hear that a lot. But NRW specifically has a well thought out funding ecosystem for SMEs. Most consultants know either the technology or the funding options, not the two combined. Integrating AI Design Sprint methodology with strategic funding planning, and the PoC and MVP implementation that follows, is the approach that moves many companies further forward than a plain consulting package.

While other companies stick to the old "We paid for everything ourselves!" approach, NRW has an ecosystem ready and waiting that makes a structured start into AI possible. Especially for SMEs. Especially now.

Why Wuppertal as a base for AI workshops

My base in Wuppertal isn't an accident. Wuppertal sits at the heart of the Bergisches Land, a region built on SMEs and industry. Tool making, metalworking, automotive suppliers, mechanical engineering. Exactly the sectors where AI creates real value.

And the location works. Düsseldorf is 30 minutes away. Cologne, 40 minutes. Essen and Dortmund, both under an hour. Whether on-site with a client, in the Bergisches Land, or at a central spot in the Rhineland or the Ruhr area, short distances are part of the offer. No consultant flying in from Munich who doesn't know your region.

I've been to the Bergische Expo, I show up at agile.ruhr in Essen, I run events in Wuppertal, and I work with companies across the Bergisches Land and beyond. The region is my home. Not my market.

Standard tools solve standard problems

ChatGPT and Gemini are quick to get and cheap to start with. But they don't know your data, don't understand your processes, and only work with what you type into the chat. Your production has its own specific challenges. Your injection moulding machine needs different settings when humidity changes. ChatGPT doesn't know that.

Nobody cares that your social media intern is writing blog posts with Perplexity. The AI in a business that actually makes a difference probably doesn't have a chat interface at all. It sits in quality control, in quotation preparation, in production planning. And getting there doesn't take 200 slides from McKinsey. It takes a workshop that starts with the real problem.

Doing is like wanting, only more intense

Stop planning. Start doing. What you don't need: a 200-slide strategy deck, endless workshops with no clear goal, checklists that keep consultants busy. What you need: one use case, one working PoC, in two to three weeks.

Cross-functional teams are non-negotiable here. A wide range of knowledge in the room: manufacturing, IT, sales, leadership. That's the only way you get solutions that actually work day to day. The AI Design Sprint brings exactly these people together.

Most AI projects don't fail because of the technology. They fail because nobody makes a decision. Because everything turns into design by committee. Because the date of the next meeting matters more than the outcome of the last one.

There are thousands of companies across NRW that could be using AI. They have the data, the processes, the problems. What they're missing is someone to sit down with them and structure the process. On-site, not over Zoom from a different time zone.

Ready to tackle your first AI project the structured way?

Back to overview