The role
The School of Computer Science is seeking a Senior Research Associate to work on the use of logic and programming languages for AI safety. The post forms part of a research project led by Dr Alex Kavvos and funded by the Advanced Research and Invention Agency (ARIA) under its Safeguarded AI programme.
This is a full-time, Grade J, Pathway 2 (Research) role. The post is open-ended with fixed funding to 30 November 2027. It is based in the School of Computer Science, Faculty of Science and Engineering, at the Merchant Venturers Building, Bristol.
What will you be doing?
You will conduct high-quality research on logic and programming languages for AI safety, contributing original ideas and developing the theoretical and methodological aspects of the project. The role includes developing formal foundations, languages and methods for specifying and reasoning rigorously about the behaviour of software and AI systems, drawing on areas such as programming language theory, logic and category theory.
Where appropriate, you will also design, implement and evaluate software tools that put these methods into practice, including through the use of agentic programming tools. You will write up research for publication in international venues, present findings at conferences and seminars, and collaborate with colleagues in the School of Computer Science and the wider Safeguarded AI research community.
You should apply if
You have a PhD, are near completion, or have equivalent research experience in computer science, mathematics, or a related discipline. You should have a strong research background in at least one of programming language theory, logic, category theory, formal verification, or a closely related area, evidenced by research outputs or equivalent professional experience.
You will have significant experience of independent research, strong analytical skills, and the ability to apply rigorous formal reasoning. You should be able to communicate complex theoretical and technical information clearly, prepare research for publication, and present findings to specialist and non-specialist audiences. Experience with formal verification, type theory, proof assistants, session types, AI safety, or agentic programming tools would be desirable.