Research Associate in Machine Learning -Generation Hardware

Overview

As AI systems scale, their energy consumption is skyrocketing. To tackle this crisis, we need to move beyond traditional computing and look at nanomagnetic devices, which offer unique, ultra‑low‑energy properties perfect for creating novel, brain‑like hardware neural networks.

We have an exciting opportunity to join the School of Computer Science as a Research Associate for an EPSRC‑funded project. You will be part of an interdisciplinary team bridging the gap between machine learning and materials science to develop next‑generation computing hardware based on nanoscale magnetic systems. This project aims to explore how systems with complementary properties can be combined to overcome the current limitations of individual elements.

In this role, you will utilise diffusion‑based generative models to simulate experimental devices and how they can be combined into heterogeneous networks. These models will allow us to use inverse design techniques to optimise network composition and train them to solve challenging real‑world tasks, such as smart prosthetics or brain‑computer interfaces.

We are looking for someone with a background in either machine learning or computational modelling and strong interest in developing novel, unconventional computing systems to tackle complex machine learning tasks. Successful candidates will contribute to ground‑breaking research that has the potential to significantly reduce the energy consumption of AI systems and accelerate advancements in the field.

• Utilise and expand diffusion‑based generative models to simulate a range of nanomagnetic systems within the framework previously developed within the group.
• Develop methods for computing task‑independent metrics and properties of neuromorphic systems to determine potential components for networks.
• Explore how devices can be combined as heterogeneous neural networks with advanced computational properties and deploy them on challenging real‑world tasks, such as brain‑computer interfaces or smart prosthetics.
• Collaborate with the project team (academics and fellow research associate) and partners to train models of physical systems and evaluate their properties over a range of task‑independent qualities.
• Communicate research findings at the local and international levels through presentations and publications.
• Engage with the research community within the University, including the Centre for Machine Intelligence.
• Keep up to date on relevant work in the field (reading and reviewing literature as appropriate).
• Work closely with the team, attend project meetings, and use collaborative tools like Google Meet, git etc.
• Carry out other duties, commensurate with the grade and remit of the post.

Our diverse community of staff and students recognises the unique abilities, backgrounds, and beliefs of all. We foster a culture where everyone feels they belong and are respected. Even if your past experience doesn't match perfectly with this role's criteria, your contribution is valuable, and we encourage you to apply. Please ensure that you reference the application criteria in the application statement when you apply.

• Hold, or be close to completing, a PhD in Computer Science, Physics or a relevant discipline (or have the equivalent experience).
• Knowledge of computational modelling and machine learning techniques, with practical experience training and evaluating models.
• Proficiency in Python, or similar, and experience with modern machine learning/scientific libraries (e.g., PyTorch, Tensor Flow, Num Py, Sci Py).
• Excellent written and verbal communication skills, with a proven ability to write up research findings for high‑impact peer‑reviewed journals/conferences and present to multidisciplinary teams.
• Knowledge of engineering mathematics (linear algebra, probability, basic calculus) and the ability to learn new mathematical tools.
• Ability to work effectively as part of a multidisciplinary research team.
• Ability to manage own research workflow, organise resources, and progress work activities independently to meet project deadlines.
• Experience in modelling or developing…