Companies increasingly integrate Artificial Intelligence (AI) into their applications to stay competitive. However, the efficient and successful development and deployment of AI applications requires complex setups. Providing developers with all the required resources, applications, and services for developing and deploying AI applications without reinventing the wheel remains challenging. Therefore, we propose a domainand workflow-agnostic reference architecture (RA) for an onpremises AI Platform that supports teams throughout the entire AI lifecycle and is reusable across multiple projects. Additionally, we present an evaluation strategy to validate the RA.

As soon as Artificial Intelligence (AI) projects grow from small feasibility studies to mature projects, developers and data scientists face new challenges, such as collaboration with other developers, versioning data, or traceability of model metrics and other resulting artifacts. This paper suggests a data-centric AI project with an Active Learning (AL) loop from a developer perspective and presents ”Git Workflow for AL”: A methodology proposal to guide teams on how to structure a project and solve implementation challenges. We introduce principles for data, code, as well as automation, and present a new branching workflow. The evaluation shows that the proposed method is an enabler for fulfilling established best practices.