Research
My research focuses on the intersection of Generative AI, Large Language Models, Prompt Engineering, Requirements Engineering, and Software Engineering.
Intelligent Software Engineering
My research investigates the application of Generative Artificial Intelligence and Large Language Models to software engineering problems, with a particular focus on Requirements Engineering.
I am interested in understanding how prompt-based and language-model-based approaches can support activities such as requirements analysis, classification, elicitation, specification, quality assessment, and knowledge extraction.
A central direction of my research is the development of reliable, interpretable, reproducible, and practically useful approaches for integrating LLMs into software engineering workflows.
Current Research Focus
- LLM-based Requirements Engineering
- Prompt Engineering for Software Engineering
- Generative AI for Requirements Analysis
- Reliability and hallucination mitigation
- Knowledge-grounded LLM systems
- Evaluation of AI-assisted RE approaches
Areas of Research
My research interests span multiple interconnected areas of artificial intelligence and software engineering.
Generative AI
Investigating the use of generative artificial intelligence for automating, augmenting, and supporting software engineering activities, particularly knowledge-intensive tasks.
Large Language Models
Studying LLM capabilities, limitations, evaluation, adaptation, and practical use within software engineering and requirements engineering contexts.
Prompt Engineering
Designing and evaluating prompting strategies for improving LLM performance on complex software engineering and requirements-related tasks.
Requirements Engineering
Exploring AI-assisted requirements elicitation, analysis, classification, quality assessment, specification, and knowledge management.
Software Engineering
Studying intelligent approaches for improving software development processes, engineering practices, software knowledge, and developer productivity.
AI for Software Engineering
Bringing together AI, LLMs, prompting, and software engineering to investigate practical and research-oriented AI-assisted development workflows.
Generative AI for Requirements Engineering
LLMs · Prompt Engineering · Requirements Engineering
One of my main research directions is the investigation of Large Language Models and Generative AI for Requirements Engineering.
This line of research examines how prompting strategies, retrieval-based methods, domain knowledge, and model adaptation can improve the performance and reliability of AI systems on requirements-related tasks.
Particular attention is given to the balance between performance, reproducibility, computational cost, privacy, interpretability, and robustness.
Research Framework
Current Research Themes
LLM-Based Requirements Classification
Classification of functional and non-functional requirements using prompting, domain knowledge, and language models.
Ontology-Guided AI
Incorporating structured domain knowledge and ontologies into LLM-based requirements analysis.
Prompt-Based Software Engineering
Systematic design and evaluation of prompts for software engineering tasks.
Retrieval-Augmented Generation
Exploring RAG-based approaches for improving the factuality and domain relevance of LLM outputs.
Reliability and Hallucination
Identifying failure patterns and developing strategies for more reliable AI-assisted requirements engineering.
Evaluation and Reproducibility
Designing reproducible evaluation frameworks, datasets, benchmarks, and experimental settings for AI-assisted software engineering.
Methods & Technologies
My research combines empirical software engineering with modern AI and language-model-based methods.
Large Language Models
LLMs and smaller language models for requirements analysis, classification, generation, and reasoning.
Prompt Engineering
Zero-shot, few-shot, chain-of-thought, structured prompting, and other prompting strategies.
Retrieval-Augmented Generation
Retrieval and knowledge-grounding approaches for domain-specific and evidence-based language generation.
Empirical Evaluation
Quantitative and qualitative evaluation of model performance, reliability, cost, and generalization.
Datasets & Benchmarks
Requirements datasets, benchmark construction, reproducible experiments, and comparative evaluation.
Systematic Literature Review
Evidence-based synthesis of existing research using systematic review and empirical research methodologies.
Research Vocabulary
Research Collaboration
I am interested in research collaboration in Generative AI, Large Language Models, Prompt Engineering, Requirements Engineering, and AI-assisted Software Engineering. Discussions involving datasets, empirical studies, systematic literature reviews, benchmark development, and reproducible experiments are especially welcome.