Case study: Ettie Eyre
Ettie completed her PhD in the Engineering and Physical Sciences Research Council-funded Bristol Centre for Complexity Science in 2013 in Language Games, and was also based at the University of Bristol’s Intelligent Systems Laboratory. Following her PhD Ettie spent six months as a postdoctoral researchers at Queen Mary University of London before moving to industry. Since then she has worked for a range of different start-ups, in data science, data engineering and infrastructure roles. She leads the Machine Learning infrastructure team at Cookpad, a Bristol-based tech company building a community platform for people to share recipe ideas and cooking tips. Her focus is on building and scaling machine learning functionality for the Cookpad application in production.
“The first year of my CDT training provided excellent preparation for a research pathway. Skills such as literature summarisation and group projects allowed me to enter my PhD program at full productivity, having already completed training in skills essential for PhD research. The cohort structure of my CDT provided peer groups helping both for research progression and contributing to pastoral support.
“Having remained within AI, my postgraduate training has had a direct impact on my career discipline. Research in academia within AI enabled me to work in AI research in industry. Part of my CDT training was conducted in collaboration with industry (for example the two group projects within the first year). I moved to industry six months after completing my CDT training. Having already experienced working with industrial partners, I was better informed to make this decision and I had some experience relevant for a role in industry.
“My role at Cookpad Global is heading up the machine learning infrastructure team. We have a team of 9 machine learning researchers working on R&D, whose focus is almost exclusively on research. In Machine Learning infra we are building systems to productionise methods which have been developed in the research team where we can find interesting use-cases within the Cookpad application. We collaborate with the researchers and the application teams to realise Machine Learning in product, and to stabilise and scale out machine learning systems. Machine Learning infrastructure engineers at Cookpad are typically people who have pursued postgraduate education in AI before following a software engineering route in industry.
“The CDT model embeds students in a centre where students are researching PhDs across a range of topics within the discipline. This exposes students to related topics of research, allowing them to grow as more rounded researchers. CDT training provides practical skills in the first year such as programming and infrastructure, which for AI research helps students to experiment effectively within their research.
“Technology businesses cannot compete in the modern landscape without leveraging AI. We invest in AI to enhance the Cookpad application, but we also invest effort into AI development so that our business can innovate with new products that leverage AI.
“AI is a growing discipline, and without the UKRI AI CDTs initiative we would be looking at a major skills shortage in UK businesses. New talent is essential for nurturing the next generation of AI academic researchers, as industry cannot thrive without an active academic community.”