Master Thesis: AI-Assisted Integrated Data Center Design for Telecom Core Systems

Join our Team

About this opportunity:
Telecom core solutions combine multiple products and cloud-native network functions. Product-level dimensioning provides inputs such as compute, memory, storage, replica and workload requirements. These outputs, together with customer requirements and engineering constraints, form the basis for the subsequent integrated Data Center design.

Integrated Data Center design is not limited to calculating or valuing hardware. It is the process of deciding how the complete solution should be structured and deployed: how many Data Centers, PODs, vPODs, clusters and node pools are needed; which functions should be placed together or separated; how storage, networking, racks, zones, failure domains and spare capacity should be configured; and how the design should support resilience, lifecycle management, operations and future expansion. 

We are looking for 1-2 students for this thesis project, based either in Kista or Lund.

What you will do:
This thesis will investigate how AI-assisted decision support can improve a representative integrated Data Center design process for core solutions. The student will combine customer needs, engineering rules and deterministic models to generate, validate and explain a complete design.

The work may include:

• Mapping the design process from customer requirements and product dimensioning; identifying the major design decisions and their dependencies across Data Centers, PODs, vPODs, clusters, node pools, CNF placement, storage, networking, racks, zones and failure domains.


• Developing a prototype that helps organise requirements, identify missing or conflicting information, and propose alternative designs; validating alternatives with deterministic checks for capacity, high availability, placement, compatibility, upgrades, resilience and physical infrastructure.


• Comparing designs using relevant criteria such as footprint, resilience, operational complexity, lifecycle impact, scalability, cost and expansion flexibility; producing traceable design documentation.


• Evaluating which activities can be automated and where human judgement remains necessary.


• Grünig and Kühn’s complex decision-making framework will provide the methodological backbone, while an agentic AI runtime will support the end-to-end process by coordinating specialised agents and maintaining the evolving problem state and traceability. 

The skills you bring:
The thesis is suitable for a student with an interest in, and preferably some experience with: 

• Telecom core networks


• Cloud-native infrastructure


• Data center design


• AI-assisted engineering


• Decision making


• Expert systems


• Agentic delegation


• Complex problem solving.