PhD by Enterprise in Manchester

Qualification Type: PhD

Location: Manchester

Funding for: UK and International

Funding amount: £21,805 per annum

Start date: September 2026

Hours: Full Time

Closes: 29 May 2026 (midnight)

The University of Manchester's PhD by Enterprise is a new four year doctoral programme that combines world class research with structured entrepreneurship training. The programme enables the University's research portfolio to generate tangible economic, environmental and societal impact through venture creation and enterprise-led pathways. The programme includes a fully funded studentship to commence in September 2026, covering tuition fees, UKRI stipend (2026/27 rate £21,805 per annum) and Research Training Support Grant.

Project details: AIDE: Agentic Intelligence for Decision-making in Investment and Enterprise. Investment and venture evaluation environments, such as venture capital, private equity, and university innovation ecosystems, are becoming increasingly data intensive. Yet despite the abundance of available information, decision-making across deal sourcing, evaluation, due diligence, and post investment monitoring remains fragmented and highly manual. Current commercial platforms excel at search and data aggregation, but they provide limited support for deeper reasoning, scenario exploration, or coordinated, lifecycle wide decision support.

This PhD project aims to address these challenges by developing next-generation AI systems capable of supporting holistic, data-driven and uncertainty-aware decision-making. Based in the prestigious Alliance Manchester Business School, the project will also explore the design and development of knowledge graphs to structure and connect heterogeneous data sources, enabling richer contextual understanding and reasoning. The project offers an exciting opportunity to work at the frontier of applied AI, decision sciences, and real-world innovation ecosystems, advancing new research while contributing to a potential future commercial venture.

A central ambition of the project is to build AI systems that are not only powerful, but also explainable. Investment decisions are high-stakes, and users must be able to understand why the system recommends particular actions or highlights certain risks. The PhD will explore explainable AI (XAI) methods that enable transparency, interpretability and user trust, ensuring that recommendations can be interrogated, justified, and adapted by human experts.

This includes surfacing the key evidence, assumptions, and uncertainties underpinning each step of the decision process, potentially leveraging knowledge graph structures to trace relationships and reasoning paths across data. The research will investigate how diverse information sources, such as structured financial data, textual documents, company disclosures, and online signals, can be integrated into unified representations that support robust reasoning, including the construction and utilisation of knowledge graphs for entity linking, relationship modelling, and semantic integration.

Equally important is modelling uncertainty: decision-makers often work with incomplete, noisy or fast-changing data. The project will examine techniques for quantifying and propagating uncertainty across multi-stage workflows, enabling users to explore how assumptions or market changes affect potential outcomes. The student will also study how multiple AI agents can collaborate to reflect real-world investment workflows, coordinating tasks such as screening, due-diligence analysis, risk assessment and scenario modelling, with knowledge graphs potentially serving as a shared structured memory and coordination layer across agents.

The design will emphasise human-AI collaboration, ensuring users retain oversight, agency, and the ability to challenge or override recommendations. Methodologically, the project blends machine learning, probabilistic modelling, multi-agent systems, explainable AI, and human-computer interaction, alongside knowledge representation and graph-based reasoning techniques. A design-science research approach will be used, with iterative prototyping, evaluation using realistic scenarios, and engagements with practitioners from investment and innovation communities.

Academic Criteria: Bachelor's (Honours) degree at 2:1 or above (or overseas equivalent); and Master's degree in a relevant cognate subject normally with an overall average of 65% or above (or equivalent) Professional qualifications and/or relevant and appropriate experience.

Desirable Criteria: A degree in Computer Science, Artificial Intelligence, Data Science, Machine Learning, Statistics, Mathematics, Engineering, Information Systems, or a closely related discipline. A Master's degree in one of the above areas. Strong analytical and programming skills (e.g., Python, machine learning frameworks) are advantageous, alongside an interest in applied AI, decision making systems, and explainable or uncertainty aware modelling. Candidates from numerate disciplines with professional experience in data science, analytics, financial technology, investment analysis, or innovation ecosystems are also encouraged. Crucially, applicants should be motivated to conduct high quality research at the intersection of AI and real world enterprise applications, with an interest in developing transparent, explainable and user centred decision support technologies.

English Language Evidence: IELTS minimum scores- 7.0 overall, 6.5 other sections. Other tests may be considered. TOEFL (internet based) test minimum scores- 100 overall, 25 in all sections. Pearson Test of English (PTE) UKVI/SELT or PTE Academic minimum scores- 76 overall, 76 in writing, 70 in other sections. To demonstrate that you have taken an undergraduate or postgraduate degree in a majority English speaking nation within the last 5 years. Other tests may be considered.

The application deadline will be 11:59PM (GMT) on 29/05/26. Apply online for 'PhD by Enterprise HUMS'. If you would like to discuss the project further, contact Prof Richard Allmendinger.