Welcome to the AI Agent Track blog

We're starting this blog to share what we're actually seeing on the ground while implementing AI agents for Dutch SMEs — not abstract AI hype, but the practical details that determine whether an implementation succeeds or quietly fails. Most failed AI rollouts we've seen didn't fail because the model was weak; they failed because no one mapped the workflow first, or because prompt quality was never treated as a real engineering discipline.

Expect posts on workflow mapping methods, prompt design patterns that hold up once real users start typing anything into them, data governance choices under GDPR/AVG, and honest breakdowns of what human-in-the-loop actually looks like day to day — not the marketing version. New posts go up twice a week. If there's a specific topic you'd like us to cover, reach out — we're always happy to write about real questions we're getting from clients.