EngD Large language models to predict and affect clean energy choices in homes

One third of worldwide CO2 emissions comes from buildings. To reach the climate goals, millions of houses need to switch to cleaner energy. This costs the residents money, time, effort, but can also help alleviate energy poverty for those struggling to pay their energy bills. Governments and housing providers search for ways to incentivize and motivate people for energy transition.

Our project develops tools to facilitate the dialog between social housing providers and their low-income tenants about energy transition, by combining large language models (LLM) with data-driven economic behaviour models. The position is part of a large transdisciplinary grant “Behaviour, Energy transition, Low income”, www.bel-tue.nl.

You will join a team of researchers and practitioners, including urban behavioural scientists (TU/e), natural language processing experts (Leiden Institute of Advanced Computer Science) and public housing providers. Your research-and-development will build upon existing behavioural models and AI- and data-driven tools of Eindhoven and Leiden universities, aiming to combine and improve them.

This EngD position is a unique opportunity that combines challenging research-and-development in close collaboration with the industry, with on-the-job training and various courses to improve your skills. You will work alternatively at TU/e and University Leiden. At completion of the project an Engineering Doctorate degree is granted.