The goal of the thesis is to design and implement a scalable, cloud-native infrastructure in Microsoft Azure that integrates automated application life-cycles with isolated multi-tenant data environments and a unified AI-assisted knowledge base.

The thesis will cover the definition of CI/CD processes and containerized deployment pipelines (via GitLab CI/CD and Azure services) to ensure seamless updates across the ecosystem. On the data layer, the work includes implementing a multi-tenant database system for private client information with dynamic provisioning and schema auto-migration. Parallel to this, a read-optimized shared global repository (e.g., a construction items catalogue) will be implemented using semantic indexing and vectorization. A cloud-based AI agent will be developed to interface exclusively with this shared global knowledge base. It will also include the implementation of centralized authentication and Single Sign-On (SSO) using Microsoft Entra ID to manage secure access to all web-based products offered by the company. The outcomes will include a fully deployed infrastructure prototype, documentation of the deployment strategies and pipeline configurations, and an evaluation of the system’s scalability, security boundaries, and cost-effectiveness. 
The infrastructure will be accessible to the thesis supervisor and the thesis opponent.