AI Compliance & Adversarial Auditing Research Assistant

1000scholars; Stafford; Ibrido; GBP 36.000 - 37.000; Part-time

Descrizione del lavoro

Keele University, School of Computer Science and Mathematics, invites applications for a part-time (0.5FTE) fixed-term role until 31 August 2027. Agile working with a blend of home and office-based work and potential for job share.

You will contribute to developing an open multimodal AI compliance and adversarial auditing benchmark, build LLM evaluation pipelines, and support reproducible experiments across text, audio, and vision-language data.

Competenze

• Master's degree in AI/ML/DS/CS/Cyber Security or closely related area.
• Strong programming skills in Python and practical ML/LLM experience.
• Desirable: PyTorch, Hugging Face, multimodal AI, LLM evaluation, cybersecurity, Git.

Mansioni

• Develop and run multimodal AI benchmark tasks across text, structured/telemetry, audio and vision-language data.
• Build and evaluate LLM evaluation pipelines; implement adversarial and robustness testing.
• Process multimodal datasets; evaluate open-weight LLMs with reproducible experiments.
• Contribute to the shared evaluation framework, documentation, open-source code, and project reports.

Conoscenze

Python

Machine Learning

LLMs

Git/GitHub

Formazione

Master's degree in AI/ML/DS/CS/Cyber Security

Strumenti

GitHub

PyTorch

Hugging Face

Reproducible pipelines

Descrizione del lavoro

Keele University, School of Computer Science and Mathematics, invites applications for a part-time (0.5FTE) fixed-term role until 31 August 2027. Agile working with a blend of home and office-based work and potential for job share.

You will contribute to developing an open multimodal AI compliance and adversarial auditing benchmark, build LLM evaluation pipelines, and support reproducible experiments across text, audio, and vision-language data.