I started by wanting to do stuff in artificial intelligence and machine learning and this was quite many years ago. I started by doing a Ph.D. in the UK at the University of Edinburgh, and then I was looking for opportunities in the same area. At that time it was not very easy now it's much easier and I ended up doing lots of different small startups. But I was not very happy with those because I didn't feel very appreciated there so I moved up to a research position at Motorola and that was good. I enjoy doing research in a big company, but Motorola was really struggling at that time, so we had some restrictions in terms of the hardware we could see people being fired, all that sort of things. It wasn't a good environment and in fact, after I left, they actually closed the whole lab where I was working. But before that, I decide to move on, and I found the job at Google in the UK. I was there for about five years and there's also flexibility at Google to move projects and I wanted a project more in tune with my interest. So I moved to Pittsburgh where they have lots of machine learning and I'm very happy here.

At Google, we are very responsible for managing our time and deciding what exactly to work on. So we are part of a project but then we have some goals for the project. So, for instance, I work in modeling PCTR and it's our goal to find different ways to improve that so getting better accuracy better results. I can manage my time exploring different ideas, collaborating with different people and trying to get there. In my weekly hours, I try to keep to the time we are supposed to do because I do have a very busy life outside work as well. I feel great flexibility in the times that I work. I think it's great from Google that I can come in and out, quite a bit of flexibility and if I have some engagement, I can go out and do that as well. But on the other end, If I have a project where I need timelines, I will do some work at home at strange hours sometimes, but I think it works well for both Google and me to have that flexibility. I don't need to travel much for work but I do travel once or twice a year, for instance, to the Mountain View or to research conference, something like that.

So at Google, we have lots of house tools and that's what I'm using. In terms of software, I code in C++, I've coded in Java. I'm doing lots of work in Python and I'm doing modeling as well, Tenserflow that's the kind of tools we use.