Data science is a highly technical profession and the skills required to become a data science hero are vast and extremely rare. This is why many large organisations are struggling to fill positions with suitable candidates.
I heard a phrase termed recently that said: "A data scientist is expensive but a bad data scientist is VERY expensive." Technical skills are of course the most vital aspect to consider when recruiting AI talent. However, essential traits are regularly overlooked or at least undervalued.
What is a trait and why are they important?
There are technical skills such as programming and statistical analysis. Playing an instrument is also a technical skill for that matter, but knowing the chords to a famous Bob Dylan track is unlikely to be useful building that sorting algorithm.
A trait, on the other hand, is a mental habit that dictates how we react to almost everything in life, personally and professionally. To come back to our guitar example, Python won't make the next Bob Dylan, but patience will prevent them from sticking the guitar in the cupboard or smashing it off the wall at the first sign of difficulty.
A common misconception is that traits are inherent and therefore set in stone. Skills and traits have one thing in common, they can be learned and they can be improved upon. It will take time but to coin a timeless phrase, old habits die hard.
So, it's all well and good in becoming a technical wizard but without the right traits, an AI professional will never be able to fully wield their data science magic powers.
Here are five essential traits of a great data scientist...
Curiosity: A great data scientist is often immensely curious about what customers want, what works, what doesn’t and why. They don’t know all the answers but by being curious, they draw out knowledge from colleagues. They use the data to prove/disprove and quantify the insights. They are truth searchers who know analytics is a journey, not a destination.
Goal Orientated: A great data scientist is not motivated by technology, tools or cool algorithms and methodologies. They are singularly focused on solving problems using data. They are focused on moving the dial, making a change and driving an impact using data and insights.
Problem Solving: Good analysts are, first and foremost, problem solvers. They like to find solutions to issues and often come from engineering or science backgrounds. They are efficient problem solvers because of their ability to take a problem, break it down into pieces, find solutions to the subparts using a hypothesis-driven approach and pull it all back together.
Clarity: A Data scientist need not only to dissect and analyze data, but they also need to explain that data to non-technical team members. A qualified data scientist can look at data and essentially tell its story, explaining how the team collected the data, how they analyzed the findings and what they predict will happen in the future. A great data scientist will be able to explain what they're doing to someone who is five or has five PhD's.
Humility: The curious data scientist knows that they don't know everything and are always looking for new things to learn. The clear-eyed data scientist swallows their pride and adapts their presentation to their audience, even if it means forgoing some ingenious technique they built. The creative data scientist thinks ‘outside the box,’ even if it feels silly. Of course, the skeptical data scientist knows to mistrust their data and their models, evaluate them all with sharp clarity and present their results with all the necessary caveats.