Master Thesis: Adaptive Test Selection using Knowledge Bases and Code Property Graphs

Join our Team

About this opportunity

Modern software is validated by long-running pipelines with large, extensive test suites. For any given change, test case selection and execution need to be optimized to run only the test cases relevant to the software changes — avoiding long waits and slow feedback. What if we could intelligently select only the tests that a change has actually impacted? At Ericsson, we want test case selection driven not just by the code, but by a richer model of the system: relating changes to an enterprise architecture model and to the relationships between Application Components, artifacts, and tests. By combining code property graphs (CPGs) with knowledge bases, this thesis explores how to turn sprawling test suites into fast, targeted, and smarter validation.​

What you will do 

This Master Thesis investigates adaptive test selection at the Application Component (AC) level across the software lifecycle. The goal is to select only the relevant tests for a given change — including integration and end-to-end tests — rather than running everything.​

You will have freedom to define areas of exploration within the project. These could include, but are not limited to:​

Extracting code property graphs (CPGs) to capture the areas impacted by a change.​

Combining CPGs with available knowledge bases (e.g. enterprise architecture models, and Application Component ↔ artifacts ↔ tests relationships).​

Evaluating how well CPGs plus knowledge bases identify the right tests.​

Building a method and tooling for test case selection, and assessing whether this approach enables faster, smarter selection.​

The skills you bring​

Knowledge in AI and Computer Science.​

Experience in at least one programming language.​

Eagerness to learn and explore new concepts