What does a data architect actually do?

9 February 2021

Andrea Palladio once led the way. As an architect, his goal was always to combine the ideal and the practical, the client’s ideas and the conditions on site, in the best possible way. When Peer Schwirtz talks about his current project and his job as a data architect at diconium, it sounds very similar. What was the Teatro Olimpico 500 years ago is the data traffic in the age of digitalisation.

“We are building a pipeline to bring data from a car to a backend and then to process it further,” says Peer, and immediately follows up with the crucial questions that drive a data architect in such a project: How many cars will be connected to the planned system? What load must it be able to handle? Which technology is suitable? What are the boundary conditions? The customer is Volkswagen. diconium has been a fully owned subsidiary of the company since January 2020.

Architecture can be agnostic

Data is the foundation of digital transformation. But it needs those who build innovations on it by making smart use of hardware and software. Data architects have a key role to play here. They provide the blueprints for a data-driven future. Unlike data engineers, for example, they are primarily strategists, developing concepts, designing data flows and orchestrating the implementation of solutions. They decide which data from which sources can be used and how.

“I started out as a software developer, but then very quickly realised that I wasn’t the best at it. What I was good at, though: Explaining things,” says Peer.  He originally studied environmental informatics and later worked for a time as a bio- and environmental informatician at the Free University of Berlin. But then Peer was drawn to the free economy – and to diconium. Six years have passed since then. As a data architect, he now breaks down the customer’s business to the technical level – and translates technical contexts for the business case. Peer is the link between the different worlds: Requirements engineering on the one hand, customer understanding on the other.

Thinking big

Palladio’s heirs talk to builders and craftsmen, the data architects to decision-makers and developers. “Management doesn’t need a technical process diagram, but the development team might want to know more precisely where the error cases are,” explains Peer. Simplifying, breaking down, abstracting – not making the complex even more complicated, that is very important and the challenge of the job. “I don’t think you have to understand so much technology. Architecture can be quite agnostic.”

What a data architect does have to do, however, is think bigger. The proverbial thinking outside the box is required. “How do different systems or components work together, for example?” says Peer. After all, no solution stands alone, but must always be seen in an environment. There are also legal questions and IT security. The data architect must keep all these levels in mind and think holistically. In the case of the Volkswagen project, for example, this also means deciding which technology can be established most quickly within the VW universe.

Do you want to become a colleague of Peer Schwirtz at diconium? Take a look at our open jobs.

Back to top