Skip to content

Blog

AI in practice: agents, automation & delivery

Hands-on guides from real projects on AI agents, n8n automation, MCP, vibe coding, RAG, and Forward Deployed Engineering.

What Is Hypercare? What Has to Happen After Go-Live of Software and AI Automation

October 7, 2026 · 11 min.

What Is Hypercare? What Has to Happen After Go-Live of Software and AI Automation

What is hypercare, how long should it last and what belongs in it? A practical guide for custom software and AI automations: reporting channel, monitoring, exit criteria and the handover to maintenance and operations.
n8n for Business: Workflows, Pricing, GDPR and Limits (2026)

September 30, 2026 · 12 min.

n8n for Business: Workflows, Pricing, GDPR and Limits (2026)

What n8n costs, how an n8n workflow with an AI agent is built, how to self-host n8n in a GDPR-compliant way, and where the tool hits its limits. With a cost calculator.
Build or Buy: When Custom Software Actually Pays Off in 2026

September 23, 2026 · 10 min.

Build or Buy: When Custom Software Actually Pays Off in 2026

Don't build everything yourself, don't buy everything either: a 3-stage model for the build-or-buy decision, backed by Gartner data, the EU Data Act, and a data-sovereignty check for SMEs.
Model Context Protocol (MCP): How Companies Chat With Their Own Data

September 16, 2026 · 12 min.

Model Context Protocol (MCP): How Companies Chat With Their Own Data

MCP is the open standard that connects AI to your data and systems – from a Google Analytics report to a legacy ERP with no API. What MCP is, how it works, the 2026 numbers, security risks, and how mid-sized companies use it well.
Bought Software Nobody Uses: Why Tools Become Shelfware in Mid-Sized Companies – and How to Prevent It

September 9, 2026 · 11 min.

Bought Software Nobody Uses: Why Tools Become Shelfware in Mid-Sized Companies – and How to Prevent It

Buying software is never the problem, using it is. Around 66% of all licences are shelfware or underused, and companies waste $21M a year on average. Why software adoption fails in mid-sized firms – with a cost calculator, a root-cause table and a clear path to real usage.
Taking Vibe-Coded Apps to Production: The Security and DevOps Check for AI-Built Software

September 2, 2026 · 12 min.

Taking Vibe-Coded Apps to Production: The Security and DevOps Check for AI-Built Software

AI-built apps get surprisingly far – until production. 45% of AI code ships with vulnerabilities, 83% of Supabase leaks are RLS errors. What actually needs checking before rollout – with a checklist, readiness check and DevOps essentials.
AI That Asks First: Why Good AI Products Ask Questions Instead of Guessing

August 26, 2026 · 9 min.

AI That Asks First: Why Good AI Products Ask Questions Instead of Guessing

When information is missing, most AI tools guess — and produce errors and hallucinations. Why an explicit clarifying-question mode and human-in-the-loop turn an AI product into a reliable one.
Custom Software for Trade Businesses: A Case Study from the Heat Pump Industry

August 12, 2026 · 9 min.

Custom Software for Trade Businesses: A Case Study from the Heat Pump Industry

Why a growing heat pump installer chose custom order-management software over Excel or off-the-shelf SaaS — real timeline, feature scope, and an honest decision framework for custom software vs. SaaS.
AI Readiness Check: The 3 Prerequisites for Successful AI Adoption

August 5, 2026 · 13 min.

AI Readiness Check: The 3 Prerequisites for Successful AI Adoption

Why 95% of generative AI pilots show no measurable business impact, and what AI readiness actually depends on: data, tools and mindset. With a self-check and current DACH figures.
Forward Deployed Engineer: What It Is and Why It Matters Now

August 3, 2026 · 14 min.

Forward Deployed Engineer: What It Is and Why It Matters Now

The Forward Deployed Engineer is one of the most sought-after roles of the AI era. Where the model comes from, how an FDE differs from consulting and freelancing, and when it pays off.
Scheduled Autonomous AI Agents: What Claude Code Routines Means for Business Automation

July 29, 2026 · 10 min.

Scheduled Autonomous AI Agents: What Claude Code Routines Means for Business Automation

Claude Code Routines shows where AI agents are heading: not just on-demand chat, but autonomous work on a schedule with real tool access. A grounded look at the RPA and cron-job comparison, the security model, and the limits.
Scrum in the AI Era: Why Sprints and User Stories Aren't the Default Answer Anymore

July 22, 2026 · 10 min.

Scrum in the AI Era: Why Sprints and User Stories Aren't the Default Answer Anymore

DORA, Scrum.org and the State of Agile Report agree: AI coding agents are changing Scrum. What that actually means for sprints, user stories, and your team.
Cutting AI Agent Costs Without Losing Quality: The Advisor and Orchestrator Pattern

July 15, 2026 · 10 min.

Cutting AI Agent Costs Without Losing Quality: The Advisor and Orchestrator Pattern

Anthropic's Advisor and Orchestrator patterns show how companies can significantly cut AI agent costs without giving up top-model quality. With verified benchmark numbers, a pricing table, and a calculator.
AI Agent Test Automation: How the /goal Command Cuts QA Costs

July 8, 2026 · 9 min.

AI Agent Test Automation: How the /goal Command Cuts QA Costs

Autonomous software testing with AI agents is going mainstream in 2026. How Claude Code's /goal command drives end-to-end test automation, what agentic QA means for MVP speed and cost, and where humans still need to review.
What an MVP really costs in 2026 — and how funding + AI agents cut your share of the bill

July 1, 2026 · 11 min.

What an MVP really costs in 2026 — and how funding + AI agents cut your share of the bill

Classic MVPs cost €20,000–€180,000. An honest look at how generalist, AI-augmented development and the current Baden-Württemberg funding programs (Innovationsgutschein, ZIM) realistically lower your out-of-pocket cost — with a calculator and funding overview.
Develop 800% faster – with on-premise AI

April 25, 2025 · 10 min.

Develop 800% faster – with on-premise AI

Build a full-stack dashboard with AI and on-premise tools in just 6 hours? In this article you'll learn how to pull it off with Bolt.new, Continue.dev, and DeepSeek v3 – without cloud dependency, but with maximum control, speed, and code quality.
Will AI Soon Replace the Clean Code Developer?

April 25, 2025 · 11 min.

Will AI Soon Replace the Clean Code Developer?

Will AI replace the clean code developer? In this article you'll learn why experience, architectural knowledge, and clean code principles remain indispensable even in the age of AI, and how smart tools like the Clean Code Agent support you in learning.
Ground Truth Data in RAG Systems

April 25, 2025 · 10 min.

Ground Truth Data in RAG Systems

Ground truth data is the foundation of successful LLM applications. In this article, you'll learn why it matters so much - and how to put it to targeted use.
AI in Software Development: 26% Productivity Increase

April 25, 2025 · 7 min.

AI in Software Development: 26% Productivity Increase

GitHub Copilot & co. boost developer productivity by up to 39% — that's what new studies from MIT & Princeton show. In this article, you'll learn how AI is changing the everyday work of development, what benefits (and limits) it has, and why juniors in particular stand to gain.
Programmieren mit KI: Schritt für Schritt

April 25, 2025 · 11 min.

Programmieren mit KI: Schritt für Schritt

KI in der Programmierung gezielt einsetzen? In diesem Beitrag lernst du, wie du mit ChatGPT & Co. deine Coding-Tasks strukturierst und effizient umsetzt.
Refactoring from Python to .NET with AI

April 25, 2025 · 8 min.

Refactoring from Python to .NET with AI

Python project migrated to .NET in just 4 hours – thanks to AI tools like ChatGPT, CursorAI & Claude. Steps, learnings, and clean-code tips in a hands-on report.