Blog
Thinking on agentic AI
Not a newsletter. Not a content calendar. Articles when there is something worth saying.
AI in the Trades: How a Heating Company Prepares Quotes in 20 Minutes Instead of Three Hours
A heating and plumbing company in Bavaria with 15 employees cut quote preparation from about three hours to 20 minutes. What was built, in which order, and what other businesses can take from it.
Bavarian Digitalbonus 2026: How to Fund an AI Project With up to €30,000
Who gets the Digitalbonus Bayern, what it funds, how the application works and which mistakes cost you the grant. As of October 2026, checked against the official programme pages.
The Company Brain: Why Your Business Knowledge Should Not Live Inside the AI Model
A company brain keeps a business's knowledge in its own knowledge base, separate from the AI model. What exactly that is, why the separation makes the difference and how to get started.
AI Process Automation in Manufacturing: Beyond the Production Line
Manufacturing companies automate production but leave back-office and quality processes manual. Here is where AI agents deliver the highest ROI in manufacturing operations — and it is not where you might expect.
The 70/30 Problem: Why Your Automation Only Handles the Easy Part
Most automation projects automate the 70% that was already manageable and leave the 30% of exceptions that consume 80% of your team's time. Here is why — and what to do about it.
What 14 Years of Enterprise IT Taught Me About Process Automation
I have spent 14 years in enterprise IT at a DAX-listed corporation. Here is what I learned about why automation projects fail, what actually makes processes work, and why I founded Jevolution to build AI agents.
What is Agentic Process Automation?
A clear definition of Agentic Process Automation — how AI agents differ from scripts and RPA, why process understanding matters more than model sophistication, and when this approach makes sense for your business.
Personal Context Protocol — Your AI Agent Shouldn't Own Your Soul
Agentic AI needs deep personal context to be useful. But who owns that context? A proposal for PCP — a protocol where humans own their data and agents just rent it.
What Agentic AI Actually Means
A clear, practical definition of agentic AI — what makes an AI agent different from rule-based automation, and why the distinction matters for business decisions.
Why Your Business Process Automation Failed
An honest look at why RPA and early automation projects lose adoption — over-rigid design, ignored edge cases, no escalation path — and what the agentic approach fixes structurally.