Glossary AI technique
Data engineering
What is data engineering?
Data engineering is the discipline of extracting data from your systems, cleaning it, combining it and making it reliably available, so reports, analyses and AI applications can run on it without manual work.
A data engineer builds the pipes that data flows through, from your CRM, accounts and ERP to the place where it is used. Those pipes are called data pipelines: they extract data, align formats and definitions, check its quality and write the result away, again and again, without anyone having to run an export. It is the groundwork of the data foundation for AI, giving every AI application a single source of truth to work from.
The classic destination is a data warehouse or data lakehouse for business intelligence. With AI, the work shifts. An AI agent needs current data from the source systems themselves, with fixed checks on what it reads. Data engineering then is as much about integrations and checks as about storage.
For EVOKE we built a uniform data layer on AFAS in this way, with automatic checks that record deviations down to individual journal entries. The AI layer that answers questions about the balance sheet, profit and loss and cash flow in plain language runs on top of it. How AI then works across your connected systems is covered under the AI Operating System.
Last updated: 20 October 2026