Touching base with: Ying Liu
“I love the excitement and flexibility of working on research and development projects”: Meet Ying, TG0’s Head of AI Research and Development
Hi Ying! Where are you from?
I’m originally from China. I first came to the UK in 2010 to study for a Master’s degree in materials at Loughborough University. I returned to China to work for a few years, and then came back to the UK to do a PhD in condensed matter and materials physics at St Mary’s University in London. That was in 2016. I’ve been here ever since.
What did you want to do as a career when you were a child?
I always loved maths and physics but I didn’t know what I wanted to do – I think I was a bit naive and was just open to whatever came along. In 2014, I was really struggling. I was working in Yangzhou as an office manager at a startup but really wasn’t enjoying it. That was when I decided to apply to study for a PhD.
Why did you want to study in the UK?
I was studying for an undergraduate degree in material science and engineering in China, and an academic from Loughborough came to my university to give a lecture. He encouraged us to consider studying materials in the UK, and a group of us came over. Many went on to do PhDs. I’d estimate a third of that group still works in the UK, a third went to the US and the rest went back to China. When I decided to do a PhD, I knew I wanted to do it back in the UK.
Have you always been interested in coding and machine learning?
I didn't really know anything about coding while I was doing my undergraduate degree – and what I did know, I didn’t like. I thought it was too complicated. I remember applying for my PhD and telling St Mary’s head of school that I was going to choose a subject that didn’t have anything to do with coding. But inevitably, while doing my experimental research, I realised that I needed to write my own software to get the insights that I wanted. That meant that I needed to learn how to code.
I started to learn Python and found it really addictive. It’s easy to pick up, and easy to understand. I went along to an event at the Alan Turing Institute in 2016. The speakers talked about how the science of machine learning was moving so quickly that those who didn’t get involved now would be left behind. I started to take online courses in machine learning on Coursera and Kaggle, and began talking to data scientists around me. And I looked out for internships or part time opportunities to get some practical experience while in the third year of my PhD.
How did you start at TG0?
I worked as a machine learning engineer for eight months, designing machine learning models of gesture classification and coordination detection for sensing electronics. At the end of my contract, I wrote my thesis and finished my PhD, but I came back to TG0 after that. An opportunity had come up to work with the company alongside the University of Leicester as part of a knowledge transfer programme, sponsored by Innovate UK. It lasted for two years and was the start of the development of TG0’s etee micro controller. We were developing a model that would customise gesture recognition for individual users.
You were promoted to Head of AI in 2022 – tell us more about that role?
I keep in close contact with different universities and am always looking for opportunities to collaborate. It’s often easier for me to be the first port of call for them because I’m quite familiar with how these studies are run and where we could collaborate on specific topics. Within TG0, the main project I’m working on is around pose estimation. I’m trying to convert the pressure sensed by our smart mat into the body skeleton – effectively turning 2D pressure into 3D body movements. It’s a big project.
What do you love about working at TG0?
I love the excitement and flexibility of working on research and development projects. There are always all sorts of new ideas coming up. There are no limitations – it’s not like we’re restricted to one type of microcontroller, or one language. If there is something we want to research or work on, we can. Everything is challenging, and everything needs R&D. I love that.
Do you have any hobbies outside of work?
I have lots of hobbies. I like doing activities after work, rather than just going home. I’m a big theatre fan – Phantom of the Opera and The Lion King are my favourites – and I enjoy hiking and rock climbing. I’ve hiked in Europe, South Africa and the US, but the Lake District is one of my favourite spots in the UK. Travelling is a big passion of mine too. This summer, I spent two months travelling around Europe by train, visiting Helsinki, Istanbul, Budapest, and Paris, to name a few. And of course, I try to get back to China once a year to visit my family.
What does AI mean for the future of product development?
AI will definitely help with hardware design, but I don’t think it will replace everyone’s jobs. Instead, it will be able to generate ideas, with experts handling the detailed work. The same goes for the software side. There will always be problems that cannot be solved by AI, and there will always be stupid mistakes made by technology. We still need people with experience to monitor it. I’m happy I’ve found my niche here. I’m very passionate about the topic and find it very interesting because it’s evolving so fast. There’s always something fresh and new to learn.
Tell us something your colleagues wouldn’t know about you
I’m a jazz singer in a band. It’s just for fun but we practise together every week. I was doing a dance course in Tower Hamlets and saw jazz singing classes advertised. I loved it and quit my dance class to sing instead. I love how it’s almost like every part of the music is showing off in some way but still together – the piano, the singing, all of these different types of instruments. It’s my favourite way to unwind after work.
If you’d like to meet Ying to ask about jazz singing, hiking or AI in product design, email her at ying@tg0.co.uk.
