PhD Candidate in Resource-Efficient and Sustainable AI

Job summary

This position offers a PhD opportunity focused on the development of sustainable and resource-efficient AI systems, particularly in the context of edge datacentres. The role involves conducting research on carbon-aware neural architecture search, enabling AI models to adapt to varying grid conditions.

Qualifications

• MSc degree in Computer Science, Electrical Engineering, Machine Learning, Applied Mathematics, or a related field.
• Strong academic background and interest in AI systems, embedded intelligence, edge computing, or machine learning.
• Strong analytical and mathematical problem-solving skills.
• Experience with programming and experimental evaluation of computing systems.
• Excellent communication skills in English.

Responsibilities

• Conduct research on sustainable AI systems for edge datacentres.
• Create hardware-aware search spaces for various computing architectures and develop multi-objective neural architecture search.
• Deliver a carbon-aware NAS framework that generates optimal model architectures and libraries.
• Publish and disseminate research findings in international conferences and journals.
• Contribute to departmental service work, including teaching assistance.
• Engage in collaborative research with academic and industrial partners.

Education

• Candidates must not hold a doctoral degree at the time of recruitment and must meet the standard MSCA eligibility criteria.
• A two-year master's degree (120 ECTS points) or equivalent is required.

Tools

• Familiarity with programming languages and tools relevant to AI and embedded systems is expected.