"PhD Position: Predicting the Energy Consumption of AI Algorithms"

Job Summary

As a PhD Candidate, you will develop innovative methods for predicting and reducing the energy consumption of large-scale AI systems during their design phase to promote environmentally sustainable AI technologies.

Responsibilities

Develop methods to assess different designs of AI systems using data flow diagrams to evaluate environmental impact. Perform measurements of AI algorithms to determine energy consumption. Predict energy scalability using generated data sets and tools like the Software Energy Lab. Calculate theoretical maximum energy efficiency for specific architectures and optimize AI algorithms. Leverage existing approaches like SaC for code generation across multiple architectures. Collaborate with a multidisciplinary team including scientific staff, PhD candidates, and software developers on AI prototypes and infrastructure.

Qualifications and Requirements

Hold a Master's degree in Computing Science or Information Science (or a closely related field). Possess knowledge of and interest in security and privacy. Be eager to work in a multidisciplinary setting with AI experts, social scientists, schools, and commercial partners. Show interest in applied research that ensures practical and sustainable contributions.

What the Employer Offers

A temporary employment contract of 1.0 FTE for 1.5 years, extendable to 4 years upon positive evaluation. Starting salary of €3,059 gross per month based on a 38-hour workweek, increasing to €3,881 in the fourth year. An 8% holiday allowance and an 8.3% end-of-year bonus. Options for 30 or 41 days of annual leave. Flexible working hours, various leave arrangements, working from home, and opportunities to compose employment conditions such as exchanging income for extra leave days. Access to training and development schemes, a good pension plan, and support for work-life balance.