Research Fellow – AI Safety and Cybersecurity

We are seeking a highly motivated and experienced 1 year contract Research Fellow to join our research team and contribute to advanced research in Agentic AI, AI Safety, Cybersecurity, Software Vulnerability Detection, and Malware Analysis.

The successful candidate will be responsible for conducting independent and collaborative research, developing novel AI-driven methodologies, and designing, implementing, and evaluating innovative solutions for software security analysis, vulnerability detection, malicious behaviour identification, and AI model safety assessment.

The role involves advancing the state of the art in AI and cybersecurity through research experimentation, prototype development, scientific evaluation, and publication of research findings.

Key Responsibilities

Research and Development

• Conduct advanced research in Agentic AI, Large Language Models (LLMs), Generative AI, AI Safety, and AI-driven Cybersecurity.
• Investigate and develop novel AI-based methodologies, algorithms, and frameworks for software vulnerability detection, malware analysis, and security risk assessment.
• Design and develop intelligent AI agents capable of autonomous task planning, contextual reasoning, tool utilisation, and complex security analysis.
• Explore innovative approaches to integrating LLMs and Agentic AI with static analysis and other software security tools.
• Conduct research on AI-assisted source code analysis, particularly for C/C++ applications, large-scale open-source software projects, and Linux-based systems.

AI Safety and Cybersecurity Research

• Develop and evaluate AI-powered techniques for identifying software vulnerabilities, malicious code, and suspicious behaviours.
• Investigate malicious activities within open-source software packages, including npm and PyPI ecosystems.
• Research advanced detection methods for credential theft, data exfiltration, malicious downloads, remote payload execution, backdoors, and code obfuscation.
• Conduct research on the safety, security, robustness, and reliability of AI and multimodal AI models.
• Investigate AI attack and defence methodologies, including adversarial techniques, model vulnerabilities, and safety evaluation frameworks.
• Design and implement experimental frameworks and benchmarks to assess AI model performance, security risks, and robustness.

Research Experimentation and Evaluation

• Formulate research hypotheses, design experiments, and conduct systematic evaluations of proposed AI methodologies.
• Develop research prototypes, experimental tools, and automated evaluation pipelines to validate research findings.
• Prepare and analyse datasets, experimental results, and performance metrics to assess the effectiveness of proposed solutions.
• Conduct comparative studies and benchmark proposed approaches against existing state-of-the-art methods.
• Analyse experimental outcomes, identify research limitations, and propose improvements to existing methodologies.

Research Publications and Collaboration

• Publish research findings in reputable international conferences, journals, and other recognised academic publications.
• Prepare research papers, technical reports, research proposals, and presentations to communicate research outcomes.
• Collaborate with researchers, engineers, academic partners, and industry stakeholders on interdisciplinary research projects.
• Contribute to research project planning, technical discussions, and the development of new research directions.
• Mentor and provide technical guidance to junior researchers, research assistants, and students where applicable.
• Support the development, integration, validation, and demonstration of research prototypes and AI-based solutions.

Technical Requirements

• PhD in Computer Science, Artificial Intelligence, Machine Learning, Software Engineering, Cybersecurity, or a closely related discipline.
• Research publications in reputable international conferences or peer-reviewed journals related to AI, Machine Learning, Cybersecurity, or Software Engineering.
• Experience conducting research in AI Safety, Agentic AI, software vulnerability detection, or malware analysis.
• Experience analysing C/C++ source code, large open-source software projects, or software package ecosystems such as npm and PyPI.
• Familiarity with static analysis tools, program analysis, software security testing, or vulnerability assessment methodologies.
• Experience developing and evaluating multimodal AI models, adversarial AI techniques, or AI safety evaluation frameworks.
• Experience with AI research frameworks and tools such as PyTorch, TensorFlow, Hugging Face, or other relevant technologies.
• Experience contributing to collaborative research projects, research proposals, or industry-academic research initiatives.