Postdoc in Agentic AI Safety & Robustness

University of Oxford; Oxford; On-site; GBP 39,000 - 48,000; Full time

Benefits offered by this job

Athena Swan Bronze award

Job summary

University of Oxford's Torr Vision Group in the Department of Engineering Science invites applications for a full-time Postdoctoral Research Assistant in Machine Learning and AI safety. The role centers on safety, control, and capabilities of autonomous AI systems that interact with tools, APIs, and environments.

The candidate will contribute to agentic safety, robustness, and system evaluation, with a background in machine learning or security and experience with LLMs or agent-based models.

Qualifications

• PhD or near completion in Computer Science, AI, Security or related field.
• Strong background in machine learning and/or computer security; experience with LLMs or agent-based systems.
• Experience with safety, security or robustness of AI systems is desirable.

Responsibilities

• Investigate safety, security, control and capabilities of advanced AI systems.
• Study and develop agentic safety, adversarial robustness, and AI control approaches.

Skills

Machine Learning

Agent-based systems

Security

LLMs

Education

PhD in Computer Science / AI / Security or related field

Job description

University of Oxford's Torr Vision Group in the Department of Engineering Science invites applications for a full-time Postdoctoral Research Assistant in Machine Learning and AI safety. The role centers on safety, control, and capabilities of autonomous AI systems that interact with tools, APIs, and environments.

The candidate will contribute to agentic safety, robustness, and system evaluation, with a background in machine learning or security and experience with LLMs or agent-based models.