About the role
We wish to appoint a Research Assistant to contribute to a UCL-sponsored research project which aims to develop models to forecast energy demand at UCL estates to cut costs and emissions. The project is a collaboration between UCL Experimental Psychology, UCL Computer Science, and UCL The Bartlett Centre for Advanced Spatial Analysis. The research assistant will be jointly supervised by Professor Maarten Speekenbrink (UCL Experimental Psychology), Professor Benjamin Guedj (UCL Computer Science) and Dr Carlo Ciliberto (UCL Computer Science). The successful applicant will develop a predictive model of energy usage patterns across locations on UCL campus. Using advanced probabilistic machine learning techniques, they will develop a robust forecasting model of short- and medium-term energy usage across UCL estates. The model will consider seasonal and other periodic fluctuations, space characteristics ( building size, occupancy, specialist equipment, etc.) to build a normative model of past and future energy usage for different locations. The model will allow real-time monitoring and prediction, as well as intelligent anomaly detection ( higher use than expected from similar locations on UCL campus) to inform decision-making and planning. Using principles of compositional kernel methods ( Gaussian Processes), the model will decompose usage patterns into components which allow for more straightforward explanation of results and generalisation of predictions across locations.
The position is for 9 months on a full-time basis. We are looking for a person with relevant experience in probabilistic modelling of time-series data and applications to large data sets. The supervisory team will provide additional support and guidance on model development and application.
We will consider applications to work on a part-time, flexible and job share basis wherever possible. This role is eligible for hybrid working with a minimum of 60% on site. This appointment is subject to UCL Terms and Conditions of Service for Research and Professional Services Staff. Please visit for more information.
About you
The successful candidate must have a University degree (BSc/MSc/MRes) in relevant area ( machine learning, statistics, artificial intelligence). They must have a deep understanding of Gaussian process regression or other probabilistic models for time-series data and experience with major Python packages for machine learning and probabilistic modelling ( scikit-learn, tensorflow, torch, pyMC). They will have excellent Python programming skills and an ability to work both independently and collaboratively. They will be able to manage their time and work to deadlines and demonstrate excellent interpersonal, oral and written communication skills. They will have the ability to work as part of a multidisciplinary team, and collaborate with other researchers. They will have a commitment to academic research and the highest ethical and professional standards in research and education. They will be highly motivated and hard-working, and ideally have an interest in sustainability and energy reduction.
What we offer
As well as the exciting opportunities this role presents, we also offer some great benefits some of which are below:
1. 41 Days holiday (27 days annual leave 8 bank holiday and 6 closure days)
2. Additional 5 days’ annual leave purchase scheme
3. Defined benefit career average revalued earnings pension scheme (CARE)
4. Cycle to work scheme and season ticket loan
5. Immigration loan
6. Relocation scheme for certain posts
7. On-Site nursery
8. On-site gym
9. Enhanced maternity, paternity and adoption pay
10. Employee assistance programme: Staff Support Service
11. Discounted medical insurance