We say we want intelligent systems.

What we really mean is—faster delivery.

“Intelligence” is treated as something we can bolt on—often via AI—rather than something we must design and embody from the start.

Despite domain-driven design, sophisticated architectures, continuous delivery, advanced tooling, and ever-more automation, most organizations still produce pipelines and documentation, not asynchronous, meaning-driven information systems.

Then everything fast becomes slow again.

This keynote argues a simple but uncomfortable truth:

Intelligence does not emerge from data, tools, or LLMs alone. It emerges from knowledge flow.

Knowledge Flow reframes intelligence as a sociotechnical property: the infrastructure that shapes how meaning is created, connected, reused, challenged, and evolved across people and technology systems.

When knowledge doesn’t flow, organizations don’t think.

They react—full of sound and fury, signifying nothing.

This is not a talk about knowledge management. This is not a talk about artificial intelligence.

This is a talk about: • our delusions about what knowledge is and who owns it • the practices that generate actual intelligence, across people and technology systems • how Domain-Driven Design enables—and sometimes constrains—the flow of knowledge

This is a talk about intelligent systems—and how architects can finally start designing them.