Research Fellow – Machine Learning for Nanoscale Semiconductor Manufacturing

Overview

Semiconductor fabrication is one of the most complex and precision-driven forms of manufacturing. At nanometre scales, even subtle variations in process conditions can introduce defects that degrade device performance, reduce yield, and drive up production costs. Addressing this challenge requires new modelling approaches that can capture the full complexity of fabrication processes and enable optimisation before physical manufacturing begins.

This project aims to develop advanced deep learning models capable of predicting fabrication outcomes and guiding fabrication recipe optimisation. By learning directly from experimental and process data, these models will enable a shift from iterative, trial-and-error fabrication towards predictive and data-driven manufacturing.

Role

We are seeking a highly motivated Machine Learning Researcher to join a multidisciplinary team of fabrication engineers and AI specialists at the University of Southampton, within the School of Electronics and Computer Science, working in the group of Dr Yasir Noori.

In this role, you will work at the interface of machine learning and semiconductor engineering, developing models that predict post-fabrication device characteristics from process parameters. You will engage with complex, high-dimensional datasets derived from real fabrication workflows, including microscopy, spectroscopy, and electrical performance measurements. You will work closely with fabrication engineers to translate physical processes into machine learning models, design and train deep learning architectures, and evaluate their ability to generalise across different process conditions. The models you develop will not remain confined to the research lab, but will be validated experimentally and tested at an industrial scale in collaboration with global companies in semiconductor fabrication and electronic design automation.

Responsibilities

• Translate physical processes into machine learning models and design/train deep learning architectures.
• Engage with high-dimensional datasets from real fabrication workflows (microscopy, spectroscopy, electrical performance measurements).
• Collaborate with fabrication engineers to ensure models reflect real-world processes and evaluate generalisation across different conditions.
• Validate models experimentally and test at industrial scale in collaboration with global companies in semiconductor fabrication and EDA.
• Supervise PhD students and junior researchers and influence the research direction of the team.
• Contribute to the development of innovative technologies with a pathway to commercialisation through the spinout Deep Fabrication.
• Publish findings in leading journals and conferences.

Qualifications

• Experience designing and training deep learning architectures.
• Experience handling complex, high-dimensional datasets from fabrication workflows (microscopy, spectroscopy, electrical measurements).
• Ability to translate physical processes into machine learning models and assess generalisation across process conditions.
• Ability to collaborate with engineers and industry partners; willingness to supervise PhD students and junior researchers.
• Track record of publishing in leading journals and conferences (preferred).

This position is offered for 24 months in the first instance, with the possibility of extension for a further 12 months.

#J-18808-Ljbffr