We go inside your engineering and show you exactly where AI actually makes a difference, even where knowledge lives only in the heads of two or three people. Requests sort themselves, reports write themselves almost automatically. GDPR-compliant, processed in the EU.
Free, 30 minutes, no sales pitch.
In the pilot, we measure the time saved on one process. We check the quality too.
Illustration of the principle · example drawing
20+ years of product development and leadership in industry · 20,000+ LinkedIn followers · GDPR-first
The engineering day-to-day
Collecting figures in Excel, updating status slides in PowerPoint, writing minutes after the meeting. The actual engineering work has to wait.
We look at these time-consuming tasks with your team and assess which preparation AI can take on. Figures, statements and decisions must remain verifiable.
What matters to your development team
Time pressure, rework and tight budgets are connected. We assess an AI application against all three measures.
Pre-sort requests, flag missing information and prepare decision documents so your team can respond earlier.
How long does it take from receipt to a reviewed result? Where do delays remain?
How we work together
At every stage, you know what happens, what you receive and what it costs. You decide whether to take the next step afterwards.
30 min
You tell me where the friction is. I tell you honestly whether AI helps, and if not, that too.
Free
1 day, report in 3 to 5 days
One workshop day with your team. Result: 2 to 3 prioritized use cases with an ROI estimate, ready to present internally. "We should do something with AI" becomes a plan with a price tag.
From €1,900
4 weeks
We automate one process and compare the time it takes before and after. We check quality together. Then you decide whether to expand.
€4,900
2 to 12 weeks
From a single workflow to a multi-agent system built on your company knowledge. Final sign-off always stays with your engineer.
From €5,000
Plus €200 to €800/month
Ongoing
After go-live I remain your contact for questions, adjustments, and new ideas.
Optional
Already know which process to tackle? You can start directly with the pilot.
Training is available at any stage or as a standalone service. Explore training ↗
The initial consultation takes 30 minutes and nothing more.
Book a free initial consultationA clear starting point
One workshop day with your team. We collect your engineering pain points, check your data and systems landscape, and prioritize the 2 to 3 use cases with the best ratio of effort to impact, each with an ROI estimate. You can present the result to your management. Fixed price, one day, no project commitment.
Prioritized use cases with ROI estimate
Reality check: what data and systems do you actually have?
Fixed price from 1,900 EUR, one workshop day
Before / after
AI takes on searching, sorting and drafting. Your team reviews results and makes the decisions. Here is how a process could change.
Search drawings, emails and reports separately.
Who remembers the original decision?
Check versions and rationale yourself.
Approved documents, respecting access permissions.
AI retrieves relevant passages with sources.
Your team checks evidence and currency.
Without reliable evidence, the question stays open.
We measure time saved and check result quality in your pilot.
You already know exactly which process costs you time every week? Then you can skip the landscape map and start directly with a pilot. We take that one process and automate it. Fixed price, clear timeline, defined outcome. Then you decide whether more follows. And if the pilot shows that AI brings nothing for your case, you will know that too. Then this insight cost you exactly one pilot, not an annual budget.
Free, 15 minutes, no sales pitch.
All modules are freely combinable, depending on what you need.
Hands-on, based on your own tasks, in the language of engineers.
Six portal learning series. In-house workshops adapted to your team.
2-day workshop for engineering teams and adjacent departments. Participants learn to work with LLMs such as ChatGPT, Claude, Microsoft Copilot, and Perplexity AI. From the basics to advanced prompt techniques for immediate results. Includes concrete use cases from the engineering day-to-day.
Every participant leaves with their own prompt toolkit for daily work
Hands-on workshop for process owners and teams. What is automation? How do classic workflows differ from AI-driven automation? Participants learn to spot automation potential and apply AI deliberately, from simple workflows to intelligent multi-agent systems.
Every team leaves the workshop with at least 3 automation use cases of its own
We determine your starting point and show you the path, instead of selling you the next tool.
Get your AI Landscape Map
Systematic audit of all AI tools in the company. Which tools are employees using (ChatGPT, Copilot, others)? GDPR risks identified, licence costs analysed, tool sprawl stopped. Consolidated to 3 approved enterprise solutions with central management.
Tool sprawl stopped, a GDPR-compliant setup with clear cost control
2-day decision workshop. Microsoft Copilot, Google Workspace AI, or a standalone solution? Requirements defined, 5 providers compared (cost, data protection, integration), clear recommendation with rationale. Contract negotiation supported.
A decision in 2 days instead of months of back-and-forth, with a documented rationale
We build systems that work every day, with your engineer as the final authority.
Have your first process automated
Chatbot searches 5,000 pages of technical documentation and responds within 3 seconds, with source references. Service technicians find answers without having to ask an expert.
Service technicians find answers themselves, with sources straight from the documentation
Intelligent email workflow with n8n. Incoming emails are automatically analysed, categorised (enquiry, complaint, order), enriched with CRM data, and routed to the right employee. Tickets are created automatically.
Every email categorised and routed automatically, no enquiry is left sitting
The development landscape
From the first requirement to a field complaint: explore selected tasks from our use-case catalogue. Start with the work that holds your team back.
Select an example
Check drawings against a checklist. Flag missing datums and conflicting specifications with their location.
Design engineers retain approval responsibility.
Explore tools for research and knowledge work. Training builds skills; implementing the workflow is a separate step.
12 selected examples from the catalogue, as of September 2026. These are application ideas, not 639 delivered projects. Feasibility depends on your data and process.
The KI-Landkarte workshop selects and prioritises suitable cases for your business.
More detailed use casesRoot-cause analysis, simulation data and idea evaluation: three examples from our own workshop, with actual screenshots and their current status.
An AI coach guides root-cause analysis and asks follow-up questions. The example shows a broken snap-fit on an assembly line.
The team confirms the root cause and corrective action.
Original screenshot · own projectA tool for querying and visualising your own simulation data. Shown here: von Mises stress in a sample component.
Load cases, model assumptions and approval remain the analyst’s responsibility.
Original screenshot · own projectAn optimist, a realist and a critical counterpart develop an idea and debate its weaknesses. The screenshot shows a debate assessment.
Scores support discussion. You decide which idea to pursue.
Original screenshot · own projectA look at our own development work. The examples are not customer references or evidence of a guaranteed time saving.Explore application ideasExperience how your team learns to work with AI
We are not tied to any vendor. We find what fits your situation because we know all the relevant tools and use them every day. When choosing, performance always matters, but so does whether a tool can be operated in a GDPR-compliant way. When in doubt, the EU-capable option wins.
Since February 2025, Article 4 requires demonstrable AI literacy from employees who use AI. Since August 2026, penalties are enforceable. No reason to panic, but a good reason not to push training into next year.
About us
You know the problem with consultants: some understand AI but not your development processes. Others know your world but bring someone else in for AI. With me you get both in one person.
For 20 years I worked in product development myself and led teams. I know where time is really lost in an engineering department, because I sat in exactly those places myself: in project reviews, in change loops, searching for that one report from three years ago. And AI has been my passion for years, not just since the hype. That combination is AImation: I translate between your engineering world and the AI world. You do not need an interpreter in between.
Who I am in 4 sentences
I know your processes, not from books, but because I lived them myself for 20 years.
If AI does not make sense for you, I will tell you that. Even if it costs me the project.
Built for the Mittelstand: fixed prices, short timelines, running next week instead of next quarter.
Training, consulting, implementation from one hand. One contact, one call.
For knowledge management projects, we use the U-KNOW.AI platform: answers with source citations instead of hallucinations, GDPR-compliant, available on-premises too.
Visit U-KNOW.AI
Holger Peschke is a member of KI Bundesverband e.V., the network for companies and experts in artificial intelligence in Germany.
Visit KI Bundesverband
Holger Peschke is a member of Beraterkompetenz Oberfranken e.V., a network of independent consultants in the Upper Franconia region.
Visit Beraterkompetenz OberfrankenFAQ
Answers to the most important questions
Ihre Daten werden DSGVO-konform in der EU verarbeitet und nicht zum Training von KI-Modellen verwendet. Ein Auftragsverarbeitungsvertrag (AV-Vertrag) liegt unterschriftsreif bereit. Und Projekte unter Geheimhaltung behandeln wir wie das, was sie sind: vertraulich. Welche Architektur im Einzelfall passt, klären wir vor dem Projekt, nicht währenddessen.
Ja, das ist bei uns keine Option, sondern Auswahlkriterium für jedes Tool: Verarbeitung in der EU, herstellerneutral, nachvollziehbar dokumentiert. Der AV-Vertrag liegt unterschriftsreif bereit. Wenn ein gewünschtes Werkzeug das nicht hergibt, sagen wir es und schlagen eine Alternative vor.
In der Regel ja, über Standard-Schnittstellen. Wir docken an das an, was Sie haben, statt neue Insellösungen zu bauen. Genau das prüft der Realitäts-Check der KI-Landkarte, bevor Geld in die Umsetzung fließt. Und wenn etwas nicht geht, sagen wir Ihnen das vorher.
Nein. Human in the Loop ist nicht optional: Die letzte Freigabe bleibt immer beim Ingenieur. KI übernimmt die Fleißarbeit, die Ihre Ingenieure vom Entwickeln abhält: Sortieren, Suchen, Zusammenschreiben, nicht die Entscheidungen. Für Gespräche mit dem Betriebsrat stelle ich Unterlagen bereit.
Die KI-Landkarte liefert nach einem Workshop-Tag plus 3 bis 5 Tage einen Bericht mit priorisierten Use Cases. Der Pilot ist in 4 Wochen produktiv. Schulungsinhalte sind am nächsten Arbeitstag anwendbar. Sie warten also nie länger als ein paar Tage auf ein erstes greifbares Ergebnis.
Die KI-Landkarte kostet ab 1.900 Euro Festpreis für einen Workshop-Tag. Der Pilot liegt bei 4.900 Euro für vier Wochen. Umsetzung startet ab 5.000 Euro, abhängig vom Agenten-Typ, plus 200 bis 800 Euro laufende Kosten im Monat. Sie erfahren den Preis vorher, nicht hinterher, keine ausweichende Antwort.
Eine Automatisierung folgt festen Regeln: Wenn A passiert, tue B. Ein KI-Agent kann darüber hinaus selbst recherchieren, bewerten und Vorschläge erarbeiten, etwa eine technische Anfrage lesen, den Kontext aus Ihren Systemen sammeln und einen Antwortentwurf vorlegen. Die Freigabe bleibt bei Ihrem Team. Mehr dazu auf der Seite zu KI-Agenten.
30-minute conversation. You get a concrete first step, even if we never work together afterwards.
Free, no obligation, answer from me personally.
2 minutes, no prior knowledge needed
