Serverless applications in banking environments

This thesis focuses on the design, implementation, and evaluation of an application based on the serverless architecture paradigm in the context of the banking sector. The goal of the work was to create a functional prototype of a system that enriches ATM data from various sources and provides it to mobile banking applications through an API. The theoretical part analyzes the challenges associated with serverless architecture and identifies the specifics of software development in the banking environment. Furthermore, functional and non-functional requirements for the system were defined, along with suitable design patterns for the given context. In the practical part, a system was first designed based on the defined requirements, incorporating the identified design patterns while respecting the specific demands of the banking domain. This was followed by implementation using tools such as TypeScript, AWS CDK, and SST, including a detailed description of key parts of the application and a testing section focused on unit testing. The solution was implemented in the AWS cloud environment using services such as AWS Lambda, SQS, DynamoDB, and VPC. Verification in a real production environment demonstrated that the application meets the defined functional and non-functional requirements, achieves low latency, is well-secured, and operates with high cost-efficiency. The results confirm that serverless architecture can be a suitable and effective approach even for conservative and heavily regulated environments such as the banking sector.