My research combines Artificial Intelligence, optimization and computational methods, with interests ranging from multi-objective decision-making to software and sustainable computing.
Development and analysis of methods for decision problems with multiple objectives, including evolutionary algorithms, adaptive heuristic selection and hybrid search strategies. A recurring goal is to reduce dependence on problem-specific knowledge without sacrificing solution quality.
Multi-agent architectures for collaborative and distributed problem solving. This line includes agent-based optimization, coordination among different decision-makers and social choice mechanisms to combine preferences, strategies or algorithmic recommendations.
Investigation of large and small language models as components of optimization systems. The focus includes hybrid architectures in which LLMs work together with expert heuristics, evolutionary algorithms, mathematical programming or multiple specialized agents.
Application of Artificial Intelligence methods to scheduling and resource allocation problems, especially in cloud computing environments. The studies consider graph-based workflows, heterogeneous machines and multiple criteria such as execution time, financial cost and carbon emissions.
Research in Software Engineering with an interest in problems that can be addressed using Artificial Intelligence, optimization and data analysis. This includes software testing, software supply chains, and the analysis of source code and software development processes.
Investigation of metrics, models and decision strategies for sustainable computing. The research includes carbon-emission assessment in different contexts, such as cloud scheduling and source-code execution or analysis, as well as the incorporation of these measures into optimization and decision support.