The System Architecture for Recombinant AI (SARAI) task structure provides an opportunity to advance the state of the art in Systems Engineering for AI (SE4AI) by enabling quick prototyping, combination, and reuse of data and AI software assets. Our team worked towards the development and integration of a pipeline of AI solutions within the scope of document processing and triage, which is shown relevant for the industry and also for the academia. This work provides the basis for demonstrating how parallel lines of AI development efforts and legacy works can benefit from an underlying system architecture that methodically integrated various AI solutions towards a desired outcome.
