Software, AI & digitalisation – explained in practice.
Building AI agents with TypeScript — and understanding every run
Building AI agents with TypeScript today is no slower than a no-code workflow. The difference is observability: how do you see in four weeks why a run failed? How we solved this at [apis.immo](https://apis.immo/) using [Mastra](https://mastra.ai/) — typed, traceable, production-ready.
Read articleDeploying AI agents in your company — what a hackathon without developers brought to light
AI agents rarely fail because of the model. They fail on routing: who needs to know what, how briefly, in which system? A one-day hackathon with four departments — and no developers — produced the answers: one workflow in Motion, a six-question framework, and a hard limit on volume.
Read articleVibe coding: why a prototype is not yet a product
AI-assisted development (“vibe coding”) builds in days what used to take weeks. But a working prototype is not secure, tested, performant, scalable software. What sits in between — and how to close the gap.
Read articleWhat does MVP development cost? (2026)
A rough orientation for budgets: what actually drives the cost of an MVP — and why a predictable sprint model lowers the risk.
Read articleTwo systems, one deploy: our setup for safe iteration
How a separated live and test system — combined with Claude Code and a GitHub pipeline — makes fast iteration possible without putting production at risk.
Read article