Retrieval-augmented generation and knowledge graphs solve different problems. We compare when each is appropriate, and when combining them gives better grounding for enterprise AI.
This article is in preparation. The full perspective will be published here soon. In the meantime, we are glad to discuss the topic directly.
AI pilots are easy to start and hard to scale. This article looks at why the architecture beneath AI — data models, metadata, access and lineage — determines whether AI can be trusted in production.
“AI-ready” is used loosely. We outline the practical characteristics — quality, context, ownership, accessibility and governance — that make enterprise data usable by AI systems.
Functional design gets the attention, but data often decides the timeline. A look at the data challenges that surface late in S/4HANA programs and how to address them early.
Tell us what you are working on and we will share a practical perspective.