Projects

This project proposes the creation of an Alternate Reality Game (ARG) entitled DREAMER, aimed at educating the general public about media. Designed as a playful, transmedia, and participatory experience, this game will utilize engaging mechanics (puzzles, false leads, realistic fictional content, urban or digital exploration) in order to develop participants’ critical skills regarding the media. From a narrative perspective, the game will build a fictional universe based on Namur folklore, thereby promoting the latter while linking playful content with real cultural points of interest in this geographic area.

This PhD project will provide diversity-based sampling techniques for the analysis and test of variability-aware Cyber-Physical Systems (CPSs). In particular, we will explore 1) formal modelling of the variability space of CPSs, 2) Exploring the notion of diversity, and 3) designing novel diversity-based algorithms. The contributions will be evaluated on a diversity of CPSs including space satellites.

Odoo is a highly configurable web-based ERP designed to manage various business operations such as CRM, inventory, sales, and invoicing, targeting small and medium-sized enterprises. Odoo is mainly developed in Python and JavaScript. This project will be conducted in collaboration with Odoo developers and will focus on a topic at the intersection of Continuous Integration and Software Testing.

The new Verification and Validation (V&V) course includes sessions where students write tests for small Java programs. Previously, a thesis explored a gamification approach centered on competition, using a leaderboard to compare students’ scores. This thesis aims to explore alternative gamification methods focused on student collaboration and to compare their effectiveness with the competitive approach used in the past thesis.

This project examines the methodological impacts of integrating artificial intelligence within a development team. It explores how AI alters established team processes, affects collaboration, and changes the decision-making landscape for developers. The goal is to identify both the opportunities and challenges that arise when AI technologies are adopted, focusing on process improvement, role adaptation, and best practices for efficient and effective AI-driven software development.

This project examines how software developers can best balance human expertise and AI automation in core tasks such as programming, documentation, and requirements management. By analyzing the strengths and limitations of both approaches, the project aims to establish clear criteria for deciding when to employ manual solutions versus AI-powered tools. Ultimately, the goal is to identify scenarios where collaboration between humans and AI can most effectively improve the efficiency and quality of software development.

The project aims to improve the updating of machine learning models, especially in federated learning where a central “super model” manages multiple instances. Inspired by software testing, RealM works to prevent updates that may cause faulty behavior. The project also plans a central system, like a package manager, to organize models, manage their versions, documentation, and performance tests over time.

This project aims to investigate the relationship between developer experience and socio-technical debt within software engineering teams. Socio-technical debt represents the accumulated costs and challenges arising from suboptimal decisions in both technical and social aspects of software development, such as poor code quality, lack of documentation, ineffective communication, and misaligned team practices. By providing actionable insights, this project seeks to help organizations foster better working environments, minimize socio-technical debt, and enhance the long-term sustainability and maintainability of their software projects.

The use of Large Language Models (LLMs) for automated test generation offers promising results but remains constrained by issues like hallucinations and prompt size limitations. This thesis investigates the integration of a graph-based Retrieval Augmented Generation (RAG) technique to enhance test generation within TestSpark, an IntelliJ IDEA plugin. We introduce GRACE-TG (Graph-Retrieved Augmented Contextual Enhancement for Test Generation), which constructs a graph of code entities using the Program Structure Interface (PSI) and rank nodes via a Personalized Weighted PageRank algorithm. This enables a precise selection of relevant context for LLMs while significantly reducing input size. Evaluation across 147 real-world Java bugs demonstrates that GRACE-TG reduces prompt sizes by over 97% compared to the current version of TestSpark, with equivalent or improved test coverage. These results suggest that graph-based retrieval can be a good candidate to improve test generation with LLMs.