Research Associate / Senior Research Associate / Research Fellow in Security Assurance for Agentic

The role

The Smart Internet Lab at the University of Bristol is seeking an outstanding researcher to join the newly funded AURA (Autonomous Unified RAN Architecture) project. AURA introduces autonomous AI-driven network management across critical network infrastructure, utilising AI agents that sense, decide, orchestrate, and propose configuration changes with minimal human intervention. This role focuses exclusively on addressing the critical security challenges created by the AI-native Future Networks architecture, ensuring the AI management layer remains trustworthy, secure, and resilient to compromise. The aim is to target agentic AI security assurance.

You will be part of a team addressing the security and resilience of private 5G networks deployed across critical infrastructure sectors. The successful candidate will conduct novel academic and high Technology Readiness Level (TRL) Agentic AI security assurance research to characterise threat landscapes, conduct adversarial testing, develop runtime assurance mechanisms, and enforce zero-trust principles on the autonomous AI layer.

What will you be doing?

• Identifying and documenting the threat landscape across the data ingestion, intent interface, and execution layers of the Agentic AI system.
• Simulating and demonstrating agent misuse and multi-agent coordination attack scenarios on a live partner platform to establish a concrete attack landscape.
• Designing and implementing intent-to-policy mapping to translate agent intent into verifiable formal policy constraints.
• Developing a runtime decision validation engine to evaluate proposed configuration actions against cross-layer security dependency models.
• Implementing dynamic trust scoring mechanisms based on historical agent behavior and continuous behavioral monitoring across decision patterns and network-layer security impacts.
• Developing automated mitigation strategies without human intervention, including rollback mechanisms and agent isolation protocols for compromised agents.
• Publishing research findings in high-impact journals and presenting at leading conferences such as IEEE Symposium on Security and Privacy, ACM CCS, or USENIX Security Symposium.
• Evaluating research experimentally using the project partner's agentic AI platform and the University's JOINER national open-networks experimentation infrastructure

You should apply if

• You hold a PhD or near completion or equivalent R&D experience in computer science, computer engineering, networks, or communications.
• You possess an in-depth understanding of fundamental cybersecurity concepts, threats, vulnerabilities, and AI applications in network security.
• You have demonstrated experience inAgentic AI security in 5G and Beyond networks.
• You have a working knowledge of telecom standards and specifications, such as O-RAN, 3GPP, and ETSI.
• You have programming experience, preferably using Python, within a Linux/Unix-based research or development environment.
• You have experience conducting experimental research and technical investigations using systematic, evidence-based approaches.
• For higher grades (Senior Research Associate/Research Fellow), you should demonstrate substantial postdoctoral or equivalent research experience, independent research leadership, and specialist knowledge in areas like AI/ML security, agentic AI, zero-trust frameworks, and cloud-native observability tools.

Additional information

To find out more about what it's like to work in the Faculty of Engineering, and how the Faculty supports people to achieve their potential, please see our staff blog:

https://engineering.blogs.bristol.ac.uk/category/engineering-includes-me/

Contract type: Open-ended with fixed funding until 29 February 2028

Work pattern: Full time/35 hours per week

School/Unit: School of Electrical, Electronic and Mechanical Engineering (EEME)

This advert will close at 23:59 UK time on 05/11/2026

For informal queries please contact: Dr. Rasheed Hussain, Associate Professor in Intelligent Networks Security Email: .

Our strategy and mission

We recently launched our strategy to 2030 tying together our mission, vision and values.