How Do You Design a Data Architecture That Still Fits Your Goals in Three Years?

Data platforms are evolving at high speed. New paradigms, new tools, and new concepts like Data Mesh, Lakehouse, Streaming, and Event-driven architectures keep emerging. At the same time, data volumes are growing, systems are getting more complex, and pressure from business units and management is increasing.

Decisions are often made too quickly: rushed, pragmatic, based on whatever is currently available. One more database here, a new tool there, a workaround that works for now. At some point, though, this patchwork starts to fail. Pipelines become fragile, responsibilities get blurry, and costs are harder to justify. The architecture that once made sense suddenly no longer performs reliably.

This is exactly where the Data Architecture key topic at World of Data comes in.

zwei btelligent mitarbeiter diskutieren über data & ai im büro

Experienced architects, engineers, and platform owners show how they transform this patchwork into robust data platforms. You see how modern architecture concepts work in real-world environments and how you can gradually evolve your existing landscape into a scalable, AI-ready platform.

You get real-world examples: from modernizing historically grown DWH landscapes and building Lakehouse architectures to near-real-time data products. You learn how governance, data quality, security, and cost control are implemented in practice, and which decisions stand the test under real workload.

What To Expect

Architecture Decisions With Substance
How to combine, for example, Data Warehouse, Lakehouse, Data Mesh, and Streaming in a meaningful way, and how to decide deliberately what is truly sustainable in your reality.

Cloud, Hybrid, On-Prem Put to the Test
Which setups companies actually run, which compromises they make, and how they successfully handle regulation, latency, and legacy systems.

Data Quality and Observability in Day-to-Day Operations
How teams set up lineage, monitoring, tests, and alerts so pipelines run reliably and issues become visible—before business units report them.

Architectures for AI Workloads
What changes in storage, access paths, security, and performance when LLMs, vector databases, and feature stores enter the picture, and how you can leverage them.

Step-by-Step Modernization Instead of Big Bang
How companies gradually transform historically grown DWH and BI landscapes into robust, scalable data platforms.

Your Questions, Answered.

Who should attend the Data Architecture sessions?

Data engineers, architects, platform owners, and tech-oriented decision-makers who are responsible for data platforms or want to take on that responsibility.

Is the focus more strategic or technical?

Both, with a clear emphasis on practice. You will see reference architectures, tools, and real lessons learned instead of theoretical concept slides.

How big a role does AI play?

AI is a core component, but not the only topic. You learn how to design your architecture so it can handle both classic BI workloads and future AI use cases equally well.

Data Architecture