Profile of a career in Astronomy and beyond

Sean Farrell

Sean Farrell

Data Scientist, Teradata

When did you first become interested in science and astronomy?

I've been interested in science as long as I can remember. My dad was an environmental scientist working in the mining industry and a big chunk of my childhood was spent living in some of the remotest parts of Australia. I developed an intense curiosity about the natural world, and in the mid 80's I was first introduced to astronomy when my family travelled out to the remote South Australian outback to watch Halley's comet slowly streak across the sky. My dad gave me a telescope for my next birthday, and the first time I saw Jupiter and its moons through it I was hooked.

What education and training did you have?

I was pretty good at maths and physics at high school, but didn't honestly think astronomy was a viable career path (I couldn't imagine that someone would pay me to do it!). So I studied Mechanical Engineering (with a double degree in Physics) at the University of Newcastle as I figured there would be jobs a plenty. Turns out there were, but they were all in heavy industry and most of the engineering jobs I was qualified for involved fixing heavy equipment. I did a stint at an aluminium smelter as an engineer in the maintenance department and quickly realised that I hated it, so I quit my job and decided to travel around Europe for a year. When my money ran out I had a serious think about what I wanted to do with my life, and decided to go back to uni and do a PhD in astronomy. I hadn't really studied much astronomy in my undergraduate degrees, but what I had fascinated me. I was lucky enough to stumble across a project that sounded pretty interesting at the University of New South Wales (at the Australian Defence Force Academy) and a supervisor who had plenty of funding to provide me a scholarship. So, in 2004 I moved to Canberra and started my PhD in high-energy astrophysics studying X-ray binaries (binary star systems where one star is a neutron star or black hole, which produce a lot of X-rays).

In which area of astronomy did you work?

During my career as an astronomer I studied what happens when stuff falls on to neutron stars and black holes, the collapsed remnants of massive stars that are so dense that they warp space and time around them. The environments around these stars are some of the most extreme in the Universe producing very high temperatures, extremely strong magnetic fields and intense gravitational fields. Very high-energy radiation is produced around these types of objects in the form of X-rays, so studying this emission tells us a lot about extreme physics that we can't replicate in laboratories on Earth. However, X-rays can't penetrate the Earth's atmosphere so we have to place our telescopes in space. So, most of my astronomy career was spent observing the Universe using X-ray space telescopes (although I also used a lot of other ground and space based telescopes in other wavelengths including the Hubble Space Telescope, the Australia Telescope Compact Array, the Very Large Telescope etc.)

How did your astronomy career progress?

After submitting my PhD thesis in early 2007 I moved to Toulouse in the south of France as a Postdoctoral Researcher at the CESR (a French astronomy institute). While there I worked on the European Space Agency's XMM-Newton X-ray space telescope mission, helping out with processing the data from the telescope and building catalogues of X-ray sources (doing research with these catalogues filled up the rest of my time). During this position a masters student I was supervising discovered an X-ray source that we think is a new type of 'intermediate mass' black hole, which could be one of the building blocks for the supermassive black holes found in the centres of large galaxies (and created quite a stir in the astronomy community). After two years in Toulouse I moved to the University of Leicester in the UK, again working on the XMM-Newton mission but this time as a Catalogue Scientist (more of an engineering role than pure research). After two years there I was awarded an independent research fellowship at The University of Sydney, where I spent almost four years working on hunting for more intermediate mass black holes.

How and why did you make the switch to data science?

After 11 years as an astronomer I decided that while I loved doing research I really didn't want to be a lecturer at a university. However, permanent jobs in astronomy that don't involve a large amount of teaching are pretty rare. I was also a little tired of moving overseas every few years, so I decided to leave astronomy and try my hand at a career that would provide better job security and stability (and better pay!) Also, astronomy jobs get extremely competitive when you get to this stage in your career, and although I came close a few times I wasn't getting any job offers. The common theme in my astronomy career was that I was always focussed on finding rare objects in big data sets, and it turns out that the skills I had developed in the pursuit of my research were very much in demand in the field of data science (named 'the sexiest job of the 21st century' by the Harvard Business Review). I was offered a job as a contractor working for the Federal Government as a Data Scientist (after meeting a guy at a wedding!) and spent 7 months developing additional skills and experience, before being offered a job as a Data Scientist at Teradata.

What skills do you use in your job that you developed as an astronomer?

The skills I developed as a researcher in astronomy were a fantastic preparation for my new career as a data scientist. Unlike most other scientists, astronomers aren't able to get up close and personal with the objects they study so they have to be particularly imaginative in how they design their experiments. Astronomy research is the ultimate detective story, forcing us to tease out subtle insights about the nature of the Universe from very limited information. We therefore tend to be extremely good creative problem solvers, a trait that is highly prized in data science (but is very difficult to teach). In addition, the hardware and software that astronomers use is on the bleeding edge of technology so we tend to be highly computer literate with good programming skills. We also have a good understanding of maths and statistics, which is pretty important. One of the most overlooked skills that astronomers have though is the ability to communicate effectively with a broad range of audiences. People are typically fascinated by astronomy, so we have a lot of opportunities to talk to the general public and the media about our work. We also spend a lot of time giving talks to other astronomers about our research as well as writing papers etc., so we tend to be good communicators (it helps having really pretty pictures to show!) On top of these skills common to most astronomers my time digging through astronomy catalogues honed my skills in data mining, in particular using machine-learning techniques to find 'needles in haystacks'. However, I think the most important skill scientists have is the demonstrated ability to master a complex topic in a relatively short period of time, so we have the ability to teach ourselves new skills as needed. This is pretty much what doing a PhD is all about.

What new skills have you developed since becoming a data scientist?

While my astronomy skill set ticked most of the boxes I needed for a career as a data scientist, it didn't tick all of them. I needed to learn quite a few new software packages and a few programming languages, but the biggest learning curve was how to analyse truly big data sets that are too big to work with on a single computer. For this you need to use distributed computing systems (i.e. clusters) or supercomputers. While working for the government I was tasked with exploring how to build analytical models with really big data sets that couldn't be processed on a single computer, which allowed me to develop experience using distributed computing software like Hadoop. I also learned a lot more about machine learning and predictive analytics, as well as learning how to process and analyse text data from sources like Twitter.

What do you like most about being a data scientist?

Being a data scientist is a lot less stressful than being an astronomer. I have really good job security, the hours are very reasonable (37.5 hrs a week standard), and the pay is about twice as much as I earned in my last astronomy position! The work is varied and interesting and there is no constant pressure to publish my results. There is also still lots of opportunity to travel around the world in my new career, but now I can also afford to do it in my own time. Also, while some of the projects I work on now are definitely focussed on making money, some projects truly have the potential to change people's lives for the better.

What do you miss about working in astronomy?

What I miss most about astronomy is being able to call myself an astronomer and being part of a really unique group of very talented people. The people I work with now are still very smart, but the intellectual problems aren't quite as challenging and there isn't as much scope for Earth shattering discoveries. I also don't have as much freedom to pursue whatever interesting problem takes my fancy, although I still have a fair bit of leeway to experiment with cool ideas. I think most of all though I miss that sense of wonder when you stare into space and discover something that nobody has ever seen before. That, and working with all the astronomy friends I made around the world!

See the ASA Factsheet on How to become an Astronomer for more information on life as an astronomer and more profiles of Australian Astronomers.

Back to top