Silverpond intern Ben Phillips at The Pond front door
In November 2016 Silverpond decided to offer its very first Machine Learning internship. Out of a pool of very high calibre candidates we landed on Ben Phillips, and by December we had welcomed him to home base at The Pond in Little Collins Street, as our first intern. Ben has been a calm, positive and constantly friendly face around the office and we have really appreciated what he has brought to the team, as well as his help with setup for Meetups and events, and his classic Dad jokes. After three months Ben is off out into the world, and I sat down with Ben to reflect back on what he had learned during his time at Silverpond.
Tell me about your background before you came to Silverpond.
Well I did my undergraduate in Mathematics, and that’s where I was introduced to programming as well, and I kind of fell in love with combining those two things. A little later on I started studying my Masters of Analytics at RMIT, and I was lucky enough to be employed by RMIT for six months as a Data Analyst. When that role finished, I was seen by Silverpond, and they thought I was alright!
So what attracted you to Silverpond as a company?
There’s a few things. One of them was the problems I’d get to work on. It’s not exactly a boring reporting job. There’s variety, and their problems really for me would push my skills and test my skills. I was really looking for an environment that I would be able to learn a lot [in]. And so I’ve been pretty lucky in that. The second thing that was a big draw was there’s quite a bit of freedom. Some of us do 9-5, some do 11-7, some work from home if their child is home, and it all seems to be pretty good.
I feel really lucky, I get to come into work and we get to talk and laugh and learn together, and I really value that environment.
I recall you telling me at the interview process that another thing about Silverpond that appealed was the Meetup culture as well.
Yeah. I remember seeing Andy Kitchen speak and I remember thinking, I wanna be him, or do what he’s doing at least!
Tell me a little bit about the projects you’ve been working on while you’ve been at Silverpond.
Sure. So initially I got to do a project just to get my head around Machine Learning, and the TensorFlow package. That was a project to make an algorithm that takes in images and identifies the gender of the person. I had Martin, Noon and Adel all then mentoring me and answering my technical questions. And so through that I was able to get a good base in that technology. From there I started to work on a research project, and that was to do with testing the idea of using simulated images for training neural networks, and to see if that could be a solution to the obstacle of not having many images to start with.
What’s in your mind the greatest skill you’ve gained or improved on while you’ve been in the internship?
I think the skills I have with TensorFlow regarding the Machine Learning… I mean I couldn’t have hoped to have gotten out as much of it as I have. Being able to do it and also being able to ask the experts all my questions, I think that’s put me in a very good position. It’s not something I could have picked up at uni. And even doing it on my own I think probably at the level I’m at, I probably wouldn’t have been able to gain the skills I have, on my own.
Ben cheerfully slaving away at his desk in The Pond
More broadly, what is it about Machine Learning that interests you the most?
I’ve been interested in nature-inspired algorithms ever since I found out about them. Just the idea that we have our set ways of doing things, but there are some very clever algorithms and we can model those algorithms to solve problems.
When we can combine the maths and the programming and these clever solutions, whether we come up with them or we discover them, yeah, I’m really drawn to that.
The Machine Learning that you’ve learned whilst at Silverpond, how will you be applying that next? Where would you like to go with this experience that you’ve had?
Well, before this internship, I had a limited understanding of what a neural network was, and how it could be applied. From my experience here, I can see that there’s quite a wide variety of ways to apply those techniques. Because of that, it has increased my interest in this particular algorithm, neural networks, a lot. At Silverpond I was able to get a good feel for what the consulting world is like, and I got to interact with clients and develop presentation skills and build on my technical abilities. I think my next steps will be to build on this experience until I’m really, really good, and I can do what my coworkers at Silverpond do. In the future I’d like to try to place myself in environments where I can develop my data science, machine learning and communication skills.
From all of the team here at Silverpond, thanks Ben, and we wish you all the best with your future Machine Learning endeavours!