Postdoctoral Researcher: AI Safety & GenAI Verification

University of Liverpool; West Kirby; On-site; GBP 36,000 - 48,000; Full time

Job summary

The University of Liverpool invites applications for a Postdoctoral Research Associate in the School of Computer Science and Informatics. You will join a vibrant research team focusing on the robustness, verification, evaluation and monitoring of Generative AI and Agentic AI systems.

You will contribute to methods to specify, assess, verify and monitor AI systems, with formal or statistical guarantees, and collaborate with partners across the UK and Europe on publications and tools.

Qualifications

• PhD in Computer Science or a closely related field.
• A strong publication record in one or more of: formal methods and AI, neural-symbolic AI, certification of learning, runtime monitoring or AI guardrails, software engineering for AI, AI alignment, expertise in probability and statistics.
• Experience publishing in venues such as NeurIPS, ICML, ICLR, AAAI, IJCAI, ICSE, CAV, TACAS or ASE advantageous.

Responsibilities

• Contribute to cutting-edge research on robustness, verification, evaluation and monitoring of GenAI and Agentic AI systems.
• Develop novel methods to specify, assess, verify and monitor AI systems with formal or statistical guarantees.
• Collaborate with academic and industrial partners across the UK and Europe and contribute to publications, software tools and datasets.

Skills

Programming for AI/ML research

Academic writing

Publications in top venues

Probability and statistics

Communication skills

Education

PhD in Computer Science or closely related field

Job description

The University of Liverpool invites applications for a Postdoctoral Research Associate in the School of Computer Science and Informatics. You will join a vibrant research team focusing on the robustness, verification, evaluation and monitoring of Generative AI and Agentic AI systems.

You will contribute to methods to specify, assess, verify and monitor AI systems, with formal or statistical guarantees, and collaborate with partners across the UK and Europe on publications and tools.