Moritz Pastow
PROGRAM MANAGER DIGITAL SERVICES & IIOT,
LEWA GMBH
LEWA GMBH
AI-driven
How we use AI in software development
Knowledge graph as context
An AI is only as good as the context it understands. We store your business case, architecture, and requirements in a machine-readable knowledge graph. This ensures the AI works with your specific project knowledge rather than general assumptions – and that knowledge remains within your company, even when personnel changes.
AI development environment
The language model is the engine, but the harness around it is the vehicle, the road, and the traffic rules. No one here uses an AI tool at their own discretion. Rule sets, guardrails, skills, and hooks define what data the AI sees, which tools it is permitted to use, and where it must stop – all documented and logged for every project.
Quality gates and approvals
Writing more code in less time doesn't automatically mean better code. That is why automated quality gates and human approval checkpoints are firmly embedded in our process – adhering to the same Clean Code standards we have upheld for over 15 years. No AI-generated code enters your application without being reviewed.
