Recursive Multi-Agent Systems for Data Discovery
Project ID: 2531bd1672
(You will need this ID for your application)
Research Theme: Artificial Intelligence and Robotics
Research Area(s):
Information and communications technologies
Engineering
UCL Lead department: Information Studies
Lead Supervisor: Vassilis Routsis
Project Summary:
Why this research is important
GenAI technology is moving beyond single models to multi-agent systems (MAS), networks of specialised AI agents that work together. An emerging field of research is recursive MAS, i.e. AI that can design and coordinate new agents to address complex problems. This project explores how such technology can transform how researchers discover and analyse complex social and economic data, including UK census data. The findings can help shape the next generation of national data services, improving how policymakers, researchers, and the public find and use this information.
Who you will be working with
You will join an interdisciplinary research team at UCL, working closely with experts in GenAI, information studies, and social data. You will also collaborate across institutions with teams at the UK Data Service (UKDS) to share knowledge and expertise on real-world data service challenges. These collaborations will help you build professional networks and insights into how cutting-edge AI can support national data infrastructure.
What you will be doing
You will design and develop recursive multi-agent workflows using generative AI models. Drawing on existing and emerging methodologies from literature, you will explore, adapt, and where necessary create new approaches that enable agents to generate and coordinate additional agents dynamically to handle diverse data formats, evolving schemas, and complex analytical requests. You will test these systems using open datasets and evaluate their performance, trustworthiness, and governance, always grounded in responsible innovation and responsible AI principles.
Who we are looking for
We welcome applicants from a range of quantitative and computational backgrounds, including computer science, data science, information science, statistics, or related disciplines. You must have strong analytical skills and an interest in artificial intelligence and large-scale data. A keen curiosity and willingness to work across disciplines, bridging AI technology and social science, are essential.