POSITION DETAILS

Department of Immunology and Immunotherapy, School of Infection, Inflammation and Immunology, College of Medicine and Health
Location: University of Birmingham, Edgbaston, Birmingham UK
Full time starting salary is normally in the range £33,002 to £35,608 with potential progression once in post to £39,906
Grade: 6
Part Time (11.25 hours per week)
Fixed Term contract up to April 2026
Closing date: 14th September 2025

ESSENTIAL SKILLS

• Degree/Masters in Computer Science or in relevant subject area.
• Hands-on experience with Neurosymbolic AI, including integration of neural networks with explainable AI techniques (e.g. Inductive Logic Programming [ILP], Meta-Interpretive Learning [MIL]).
• Experience working with AWS, including deployment and scaling of AI models and web applications.
• Proficiency in Python, with a solid understanding of relevant libraries (e.g. Hugging Face Transformers, PyTorch/TensorFlow, Flask/Django.
• Experience embedding LLMs into web platforms, with a user-centred approach to interface design and accessibility.

DESIRABLE SKILLS

• Understanding of the needs of stroke survivors and caregivers, with a commitment to inclusive design and ethical AI.
• Familiarity with LLM fine-tuning and prompt engineering, especially for healthcare-related conversational agents.
• Understanding of data privacy and security standards, particularly in healthcare contexts (e.g. GDPR compliance).Experience working in interdisciplinary teams, including clinicians, researchers, and patient advocates.
• 
  Further particulars can be found here
  Informal enquiries to Sandeep Shirgill, email: s.shirgill@bham.ac.uk
  We believe there is no such thing as a ‘typical’ member of University of Birmingham staff and that diversity in its many forms is a strength that underpins the exchange of ideas, innovation and debate at the heart of University life. We are committed to proactively addressing the barriers experienced by some groups in our community and are proud to hold Athena SWAN, Race Equality Charter and Disability Confident accreditations. We have an Equality Diversity and Inclusion Centre that focuses on continuously improving the University as a fair and inclusive place to work where everyone has the opportunity to succeed. We are also committed to sustainability, which is a key part of our strategy. You can find out more about our work to create a fairer university for everyone on our website

ROLE SUMMARY

We are seeking a part-time AI Developer to support the design and implementation of a Large Language Model (LLM) chatbot tailored to the needs of stroke patients and their caregivers. The successful candidate will be responsible for developing and embedding the chatbot into a web-based platform, ensuring it delivers accurate and accessible support for stroke patients and their carers. This role combines technical expertise in AI and software development with a strong commitment to inclusive design and ethical technology in healthcare. The developer will work collaboratively with the research team and industry partner to ensure the solution is both technically robust and user-centred. The role is for the duration of 10 months commencing on 1st October 2025. The work is funded by the EPSRC network, N-CODE.

MAIN DUTIES

• Support the design and deployment of a Large Language Model (LLM) chatbot tailored to the needs of stroke patients and caregivers.
• Integrate the chatbot into a secure, accessible web-based platform with a user-friendly interface.
• Apply Neurosymbolic AI techniques to enhance explainability and reliability of chatbot responses.
• Collaborate with healthcare professionals, researchers, and user representatives to ensure the chatbot meets clinical and emotional support needs.
• Fine-tune and evaluate LLM performance using real-world dialogue data and feedback.
• Ensure compliance with data protection regulations and ethical standards, particularly in healthcare contexts.
• Manage cloud infrastructure (AWS) for hosting and scaling the chatbot application.
• Monitor system performance, troubleshoot issues, and iterate based on user feedback and evolving project goals.