Opportunities
PhD and MSc by Thesis projects
Projects DAIM researchers are offering to research students. Each one names the supervisor, what the project investigates, and what you would need to bring to it.
-
Foundation Models for Engineering and Scientific Data
The project
This project explores the development of foundation models capable of learning transferable representations from complex engineering and scientific data. Building on recent advances in deep learning, self-supervised learning and time-series foundation models, the research aims to develop AI methods that can generalise across different datasets, engineering systems and prediction tasks.
Potential research topics include time-series forecasting, representation learning, multimodal data integration, transfer learning and physics-informed modelling. Applications may span energy systems, intelligent infrastructure, manufacturing and other scientific and engineering domains.
The project offers opportunities to investigate fundamental AI methodologies while addressing real-world engineering challenges, contributing to the development of more generalisable, data-efficient and trustworthy foundation models.
What you will need
A working knowledge of Python and machine/deep learning frameworks, with an interest in foundation models and scientific applications. Experience in time-series analysis, PyTorch or mathematical modelling would be advantageous.
-
Autonomous AI Agents for Engineering Decision-Making
The project
This project investigates the development of autonomous AI agents capable of reasoning, planning, learning and making decisions in complex engineering environments. By integrating large language models, reinforcement learning, optimisation and multi-agent systems, the research aims to advance AI agents that can interact with engineering models, digital environments and computational tools.
Potential research topics include reinforcement learning for sequential decision-making, LLM-based agent architectures, multi-agent coordination, tool-augmented reasoning and autonomous scientific experimentation. Applications may include energy management, intelligent infrastructure, transportation and industrial operations.
The project aims to develop reliable, adaptive and explainable AI agents that can support engineering analysis, explore alternative strategies and improve decision-making under uncertainty.
What you will need
A working knowledge of Python and machine/deep learning frameworks, with an interest in reinforcement learning, large language models or autonomous agents. Experience in optimisation, control systems or engineering simulation would be advantageous.
-
AI for spatial and temporal crime prediction
The project
This project introduces a sophisticated crime prediction framework that leverages AI method (like Graph Deep Learning, Reinforcement Learning) within a spatio-temporal big data context. By synthesising historical crime records with socio-economic and environmental datasets, the study aims to achieve high-resolution predictive accuracy across granular spatial and temporal scales. The resulting system is designed for seamless integration into law enforcement workflows to optimise public safety interventions.
What you will need
A working knowledge of Python and machine/deep learning frameworks.
-
AI for urban mobility analytics
The project
Leveraging AI methodologies is instrumental in extracting actionable insights for urban planning and public resource management. As optimising mobility is fundamental to sustainable development and equitable planning, this project investigates complex urban movement patterns through the analysis of diverse big data using advanced AI techniques, such as Graph Neural Networks (GNNs) and Large Language Models (LLMs). Examples of research focus include identifying the socio-spatial drivers of commuting behaviour, forecasting origin-destination (OD) flows, and utilising current big data to conduct demand analysis for urban planning.
What you will need
A solid foundation in Python programming and spatial data analysis. Beyond that, a strong intellectual curiosity and a drive to learn advanced AI methods to uncover deeper insights within complex urban datasets.
-
AI for sustainable and resilient cities
The project
This is a project that uses artificial intelligence to help cities optimise resources, reduce carbon emissions, and build resilience against climate-related challenges. The objectives include improving energy efficiency, predicting flood and heatwave risks, managing traffic and waste, and analysing carbon emission patterns across urban sectors such as transport, buildings, and industry to support net-zero targets.
Methods such as machine/deep learning, computer vision, reinforcement learning and LLM are applied to data from IoT sensors, satellite imagery, smart energy grids, emission monitoring stations, and socioeconomic records. The potential implications include reduced greenhouse gas emissions, lower operational costs, smarter climate-adaptive infrastructure, and more equitable urban planning, making AI a critical enabler for the transition toward sustainable and resilient cities.
What you will need
Python programming, basic machine learning and data science knowledge (e.g. scikit-learn, TensorFlow/PyTorch), familiarity with geospatial tools (e.g. QGIS, GeoPandas), understanding of data wrangling and visualisation, and a foundational understanding of urban systems, environmental science, or climate change is beneficial.
-
AI driven risk assessment in modern slavery
The project
This project proposes the development of a Human Centered AI system to proactively identify, assess, and mitigate forced labor risks within complex global supply chains. By integrating Large Language Models with Retrieval Augmented Generation and Agentic AI workflows, the system will dynamically analyse fragmented and unstructured data, such as global news, audit reports, and legal frameworks, to detect hidden signals of exploitation.
To establish a robust evaluation framework, this approach will be benchmarked against alternative methodologies. The practical efficacy of the solution will be validated through a targeted case study focusing on one industry sector to surface localised vulnerabilities. Crucially, the system moves beyond static risk assessment by functioning as a remediation engine, generating context aware, ethical, and legally sound intervention strategies.
What you will need
A strong foundation in Python programming and practical experience with machine learning and deep learning methodologies. The ideal candidate will have a proven ability to build and optimise end to end data pipelines and be highly familiar with standard artificial intelligence frameworks. Crucially, the candidate must demonstrate a genuine interest in applying this technology for social good, specifically in the area of combating modern slavery in global supply chains.
-
Democracy in the digital era: social media as a tool for assessing democratic quality
The project
This study will explore whether social media data can serve as a faster, real-time alternative to traditional democracy indices for assessing the health of democracy in India. Using AI tools, particularly open-sourced large language models like llama3, the project will analyse Reddit posts to extract key democratic variables such as trust in government, misinformation, political polarisation, and media exposure. These variables will then be correlated with established democracy indices (such as V-Dem and Freedom House) to understand how online political discourse reflects and influences a country’s democratic health. The ultimate goal is to lay the groundwork for a global early warning system capable of detecting signs of democratic decline in real time.
What you will need
Technical skills. Python programming (for data collection, processing, and ML model training); experience with machine learning and NLP techniques (text classification, sentiment analysis); familiarity with large language models and APIs; basic data preprocessing and handling of unstructured text data; comfort working in cloud-based computing environments like Google Colab.
Domain knowledge. Understanding of, or interest in, political science concepts, particularly democracy indices and electoral democracy; awareness of social media as a data source and its limitations; basic knowledge of research ethics around data privacy and anonymisation (GDPR awareness).
No project carries that keyword. Clear the filter to see all 7.
Applying
Write to the supervisor named on the project before you apply. Their contact details are on their University staff page, linked from each entry. Say what draws you to the project and what you have done that is relevant — a short, specific message is worth more than a long general one.
The application itself goes through the University. Entry requirements, fees, start dates and the form are all on the Data Science research degrees page, which covers both the MSc by Thesis and the PhD.
For the centre itself — people, facilities and the wider research programme — see the University of Hull DAIM page .