Real Talk With A Data Scientist at Edge

Hi Roberto ! Could you tell us a bit about yourself?

I was born in Venezuela and grew up on Spain. I studied Biochemistry and had a big interest in Microbiology. One thing led to another, I moved to London to do a Masters in Molecular Virology at Imperial College.

After that, I had an opportunity to do a PhD combining lab work and bioinformatics. I found it interesting to analyse large datasets that don’t mean anything at first but when analysed in the right way, provide exciting insights. It is going from gigabytes worth of data to something meaningful. For example, while analysing the data for my PhD, I saw that infants that lived with furry pets, such as cats or dogs, carried more bacteria in their throats that could protect them from developing bacterial meningitis.

After finishing my studies, I wanted to apply what I learnt to a healthcare setting, and this is how I ended up at Edge!

Finally, I am a big fan of coffee, it keeps me going!

What technologies are you working with?

R is by far the programming language that I use the most at work. We use Shiny to create interactive dashboards. It is great because our clients can directly engage with their data, rather than working with static plots and tables on slides, and view the data according to their needs.

I also use R Markdown and LaTeX to generate PDF reports for some of our clients. Combining their functionality, we can automate report generation and customize the products for the clients (ie. logo, pages).

We are also looking to expand our expertise into Tableau and Python which could help to adapt and satisfy our clients’ needs and requirements.

What is the most interesting challenge you have been working on so far and how do you deal with it?

The shift from Academia where I was, for most of the time I was trying to solve one big challenge, doing a very specific type of analysis to achieve my goal. Work outside of academia is much more collaborative and fast-paced. I get the opportunity to work across various projects simultaneously, each requiring different tools as I mentioned earlier. This can also be a challenge at first, I had to adapt and change my mindset and the way I approach the work itself. Being flexible to change is key in a start-up environment where technologies keep changing and new ones are being introduced as we grow.

As an example, not too long after I started, I was offered to chance to work on a project with one of our clients at the Sandwell and Birmingham NHS Trust. I had to learn about operating theatre scheduling and the process, understand the problems the bookers had, and try to come up with an effective solution. We ended up creating a Shiny app that helps them by suggesting optimised patient lists, allows them track patients waiting list, allows them to visualise what theatres are not used efficiently. I was a good learning experience for me! I haven’t been in a position in which I had to listen to client, learn about a completely new field of work, and think of a good way of solving their problems before.

How would you define machine learning for beginners? And how do you apply it in practice at Edge?

Machine learning has various definitions and is a topic that is constantly being debated. For example, the way we apply it at Edge is by combining algorithms and statistical methods to extract insights from the different data sources we have. For example, use historical data from surgeons, patients and anaesthetists to predict how long a procedure will take to optimise operating theatre lists. It’s exciting!