In today’s digital transformation landscape, process choreography is crucial for coordinating workflows across organizational boundaries. However, existing approaches generally assume common modeling languages, execution environments, or even modeling paradigms—constraints rarely met in practice. We present pMage, a low-code platform enabling runtime process choreography across heterogeneous and cross-paradigm environments without design-time integration. pMage acts as an integration-layer mediator that observes state changes across independent PMSs and enforces interaction rules defined at runtime, enabling choreographed interactions while preserving each system’s autonomy. To support diverse platforms, pMage combines curated mappings with lightweight NLP techniques to assist users in aligning platform-specific terminologies and identifying equivalences between process elements and execution semantics, without requiring unified models. A prototype illustrates how pMage coordinates processes in cross-platform and cross-paradigm settings, demonstrating its potential for distributed and federated process ecosystems.