Implementace datových produktů pro investiční bankovnictví

The dynamic development of investment banking brings new requirements for the development of information systems and data processing. In response, the concept of data products is coming to the forefront, allowing data to be treated as independent software components using proven engineering practices, thereby supporting decentralized data management in an agile environment. The main objective of this thesis is to design and implement a functional prototype of a data product within a selected banking institution using YAML and Python, and subsequently verify its practical benefits. To achieve this, Design Science Research (DSR) was chosen as the primary methodological framework, supplemented by a literature review and the MMSP methodology, which was used to develop a supporting web application for a central data product catalog. The result is a fully functional data product prototype, whose architecture consists of communication interfaces (Input, Output, Observability, Discovery, and Control Ports), application logic, and infrastructure. The theoretical architectural design and data contract definition were successfully validated through expert interviews with industry professionals. The implementation of the internal logic and CI/CD automation processes (sub-objectives DC3 and DC4) was verified through the triangulation of evidence, combining test scenario execution, production CI/CD log analysis, and mocking. Subsequently, the developed data catalog application (sub-objective DC5) was validated using the standard test pyramid concept, fully covering Unit, End-to-End (E2E), and User Acceptance Testing (UAT). The thesis thus fulfilled its main objective and provides a verified technical blueprint for the financial sector. Future development should focus on implementing active Data Quality Indicators (DQI), extending the data contract with a semantic layer, and integrating a vector database for advanced artificial intelligence use cases. Further research may also explore the transferability of this architecture to other industries outside of banking.