The AI-Driven Research Software Engineering (AI-RSE) initiative integrates artificial intelligence into the processes of designing, developing, and maintaining research software. Hosted by Wageningen University & Research, the initiative emphasizes creating robust, reusable tools that align with environmental responsibility and adhere to the FAIR (Findable, Accessible, Interoperable, and Reusable) principles. These principles ensure that research outputs are systematically organized, accessible, and reusable for a broad range of applications.
The AI-RSE initiative aims to enhance the efficiency of research workflows through AI-driven tools that optimize resource consumption. A significant focus is placed on developing adaptable frameworks to maintain the usability and relevance of research software over time. By addressing technical debt in software design, the initiative prioritizes sustainable practices that improve software reusability and longevity. Additionally, AI-enhanced systems are employed to support data-driven decision-making, facilitating precise and efficient knowledge management. The initiative adheres to FAIR principles to ensure that software and data are accessible, reusable, and interoperable, enabling seamless integration across research domains.
Sustainable practices are a fundamental aspect of the AI-RSE initiative. The development of lightweight and scalable software prioritizes efficient use of computational resources, reducing energy consumption. Open-source methodologies ensure transparency, collaboration, and widespread accessibility of tools, fostering community-driven support and maintenance. Automation is leveraged for software updates and monitoring, thereby extending the operational lifecycle of research software. Green AI principles guide model training and deployment to minimize environmental impact. These efforts are complemented by alignment with FAIR practices, ensuring the long-term accessibility and interoperability of outputs for diverse applications.
Current research focuses on addressing domain-specific challenges in software engineering for scientific applications. Automated extraction of requirements and problem definitions through AI facilitates the rapid translation of research needs into functional software solutions. A centralized knowledge graph is being developed to consolidate information across various software tools and technologies, providing an organized and accessible repository for research software engineering (RSE). AI-driven automation is employed to streamline software generation, deployment, and maintenance, ensuring efficiency and scalability. These efforts are aligned with FAIR principles to enable interoperability and adaptability across evolving scientific and technological landscapes.
The AI-RSE initiative welcomes collaboration with students, researchers, and software engineers who specialize in artificial intelligence, research software, and data management. This collaborative approach supports the development and refinement of tools and frameworks to advance research software engineering practices. For inquiries or opportunities to contribute, please contact:
Contact: Siamak Farshidi at siamak.farshidi@wur.nl.