Domain-Driven Design (DDD) has long been the go-to approach for modeling complex business domains, with bounded contexts defining clear boundaries. But when the primary executor shifts from humans to AI agents, those boundaries become obstacles rather than aids. Agents need to reason across contexts, requiring a shared semantic foundation. This post argues for adding an ontology layer above DDD—a formal, machine-readable specification of concepts and relationships that acts as a semantic contract. This allows agents to interpret data and rules consistently across bounded contexts, enabling more autonomous and reliable operations. The idea is still emerging, but it points to a future where enterprise architecture must accommodate both human and machine actors with explicit semantic agreements. For architects and developers, this suggests investing in ontology modeling skills and tools that bridge the gap between domain logic and AI reasoning.
As AI agents take over execution roles, DDD's bounded contexts hit their limits. This post proposes an ontology layer as a semantic contract to bridge contexts for agents.