Recently, there has been a significant breakthrough in artificial intelligence (AI) with the development of Large Language Models (LLMs). Trained on vast datasets (encompassing a wide range of knowledge from diverse sources), LLMs have shown an impressive capability to elaborate context-sensitive human-like text and respond to complex queries coherently. Moreover, LLMs have shown surprising emergent abilities in complex tasks, like instruction following, arithmetic calculations, logical reasoning, and code generation/debugging. These specialised applications of LLMs are revolutionising industries by providing accurate, timely, and context-aware solutions, promising a future where AI seamlessly integrates into various professional domains, including spatial analysis and modelling.

You will work on an innovative project entitled "IntelliGeo: Laying the Groundwork for Integrating Large Language Models into GIS." This project aims to improve geographical modelling by integrating LLMs with open-source GIS platforms. Our goal is to enable AI and human authorities to collaboratively design intricate models to address challenges such as flood detection, crop yield estimation, and prediction of environmental changes. By joining our team, you will contribute to cutting-edge research at the intersection of Geospatial Information Science (GIScience) and AI, advancing both fields and championing a European vision of human-centric AI.

The work spans from developing the basic functionalities of a QGIS plugin to prompt engineering and fine-tuning of a selected LLM based on the user-generated data gathered by the plugin. Specifically, you will be collaborating closely with the Principal Investigator (PI) and a Research Engineer (RE) to develop IntelliGeo, a plugin for QGIS that leverages LLMs for enhancing geoprocessing workflow modelling. You will also be contributing to generating a fine-tuning dataset for GIS tasks, engaging in the preliminary fine-tuning of LLMs using the generated dataset, and contributing to assessing the efficacy of the fine-tuned LLMs for GIS tasks. You will be participating in workshops and conferences to introduce IntelliGeo to the GIS community and gather feedback and will be collaborating with partners, including the Barcelona Supercomputing Center, to implement and refine LLMs.

• An MSc or PhD in a related field (e.g., Software Development, Computer Science, Geo-Information Science)
• Analytical and problem-solving skills, with the ability to work collaboratively in a multidisciplinary team
• Strong communication skills, with proficiency in English
• Programming skills in Python programming and Python ecosystem for machine learning / deep learning
• Experience in working with Natural Language Processing models is highly regarded
• Familiarity with developing plugins for GIS software, e.g., QGIS/ArcGIS, and a track record of publishing in peer-reviewed journals or presenting at international conferences are considered a bonus

• An inspiring multidisciplinary, international and academic environment. The university offers a dynamic ecosystem with enthusiastic colleagues in which internationalisation is an important part of the strategic agenda
• Full-time position for 10 months
• Gross monthly salary between € 3,226.- and € 4,036.- (job profile Researcher 4)
• A holiday allowance of 8% of the gross annual salary and a year-end bonus of 8.3%
• Excellent support for professional and personal development
• A solid pension scheme
• A total of 41 holiday days per year in case of full-time employment
• Excellent working conditions, an exciting scientific environment, and a green and lively campus.

The Department of Geo-information Processing (GIP) is a multidisciplinary scientific department that develops computational methods for processing spatiotemporal data, which in turn are used to build models, visualisations, and information systems to improve our understanding of dynamic spatial systems and to help in decision-making at multiple spatial and temporal scales. GIP is not tied to a single application domain but develops generic geo-information and geo-products that are widely applicable.