What really makes a good candidate in the data industry?
A recent article in Data IQ this week suggested that experience was valued more than qualifications in the data industry’s recruitment process. This question has sparked debate in the Starcount Offices and we got the expert opinions of a Senior Data Scientist, a Data Engineer and our Chief Product Officer on the subject. If you’re a data scientist or engineer looking to advance your career or break into the market then this one’s for you!
Data IQ – Data qualifications “not as key as experience,” say firms
New graduates looking for their first job in the data industry could struggle to get their foot in the door at many businesses after it has been revealed that most firms put practical skills ahead of academic qualifications in their recruitment process. Only 18% viewed a BA or even an MA or doctorate as a primary consideration when hiring…
Here’s what the Starcount team had to say…
Data Engineer: “I believe a balance is required here. From my experience, companies put too much emphasis on academics so it’s refreshing to see this article offering an alternative point of view. Although qualifications can be a good indicator of a candidates ability, they shouldn’t be the only consideration
I think candidates who have demonstrated ability through experience are just as valuable, particularly within software development and engineering. This experience can come through personal publications, contributions to open-source projects or their own research they’ve published online. I try to encourage anyone who asks to get involved in projects around the study.”
Senior Data Scientist: “Within Data Science, experience is highly valued but not essential. Before Starcount I attended a Science to Data Science (S2DS) course and started as a Junior Data Scientist. As a data scientist once you have a good grasp of the languages and techniques you are more-or-less ready to start work! It is not as important to think laterally as it is as an engineer.”
Data Engineer: “For all roles, I think it’s important to look at how a candidate approaches a problem. A person’s ability to understand and apply what they’ve learnt to abstract problems is paramount whether the role is engineering or statistically focused. Enthusiasm for how the world works from an empirical point of view and a desire to deeply understand how things work in general, I believe to be strong indicators of how someone will progress.
Senior Data Scientist: “At Starcount we look for data science candidates with good exposure to statistics. This is because usually, you have to deal with weak data – the ability to pick up data ‘signals’ from ‘noise’ is a valued skill as the signals are often hard to find and disentangle. Whatever your discipline it is important to be able to change and adapt your skills to the needs of the company and maintain a problem solvers mindset.”
Data Engineer: “Candidates should be able to explain how algorithms and methods work from first principles. I often see people applying learned techniques incorrectly because basic principles have been missed, the qualifications don’t guarantee a person’s ability to translate what’s taught to real-world problems. From my experience, courses tend to have data that is organised and clean, whereas in reality data is provided in many different formats and often requires a great deal of cleaning. This can sometimes give an inaccurate, polished view of what might be required.”
Chief Product Officer:
At Starcount, we have debated this concept during the hiring of multiple candidates. And the truth is, the answer isn’t binary…
When hiring new candidates, qualifications are an obvious place to start, but you can have very “qualified” candidates who tend to be too academic. We particularly find this with PhDs. They have often become used to an environment where they are seeking statistical perfection and have substantial time to investigate and prove theories. On the other hand, we see candidates with oodles of experience who have picked up bad habits and been taught reliance on software, people and business rules.
At Starcount, we find that the most successful candidates are those who are self-motivated and have a problem-solver mindset. When hiring for data scientists and engineers, we look for candidates who tackle problems laterally, enjoy working collaboratively and always ask “why?” These are the people who find success in data careers due to their constant curiosity for the right answer, their fulfilment in being part of a team and their ability to step back from a problem and answer it with common sense – neither experience nor academic prowess can substitute the right mindset.