Hackernoon logoRead This Guide Before Hiring a Data Scientist by@joey

Read This Guide Before Hiring a Data Scientist

Author profile picture

@joeyJoey Bertschler

Let's talk 🚀linkedin.com/in/joey-bertschler

Hiring the right talent for a data science role is challenging, but the answer isn’t to resort to traditional hiring practices.
Data science is more than a buzzword – it’s a booming field that promises to have one of the best global industry outlooks for many years to come.
You’ve likely heard the news that Data Scientists are the sexiest jobs of the 21st century, and while the term “Data Scientist” was coined just over a decade ago, data science has since become one of the highest sought-after professions across industries. It’s also ranked as the third-best job in America for 2020, with a median base salary of US$107,801 and a job satisfaction rating of 4/5 according to Glassdoor. To top it all off, the LinkedIn 2020 Emerging Jobs Report ranks Data Scientist third on its list, with an annual growth rate of 37 percent.
At some point along your entrepreneurial journey, you’ll likely realize that your company requires better understanding and manipulation of data that necessitates hiring a Data Scientist. And according to IBM, this need will only continue to increase across industries – they predict that the demand for Data Scientists will soar 28 percent by 2020. 
However, demand for Data Scientists is far outpacing supply. This scarcity has led to a highly competitive average salary, and likewise increased the risk of hiring with its significant costs. Because Data Scientists are such a challenging role to hire, recruiters must approach hiring them in the right way.

Before hiring a Data Scientist

To successfully hire a Data Scientist who’s right for your company, first you need a plan. Be sure you have a thorough understanding of the following:
Understand why you need a Data Scientist. What projects will they work on? What business decisions will they provide insights for? Might there be an easier way to get those insights? These are all questions you need to answer before reaching out to hopeful candidates.
Visualize who you want to hire for the role. Does the role require entry-level or more senior skills? Do you need a generalist or a specialist? Might you be better off with a Data Engineer or a Data Analyst? People hired to handle your company’s valuable data must be relevant to your projects, and these questions will help contextualize your new hire before you start recruiting.
Solidify the job content and requirements. This is true of all new hires, but particularly so for the daunting task of hiring a Data Scientist. The job description should be specific but not too demanding in nature. A professional Data Scientist will not waste time applying for a job description that asks for the impossible or unreasonable – for example, 10 years of experience in deep learning when the field has barely existed that long.
Decide where and how the new hire will fit within your organization. This is particularly crucial for startups that are laying the foundation for their future business, but also true for larger corporations. As a general principle, a Data Scientist hire should be situated within the company to make optimal use of existing resources to benefit the company’s operations.
Once a detailed plan has been established, it’s time for the easy part: follow-through. A worrisome but very real fact is that common hiring and interview processes for Data Scientists are flawed due to the challenges of verifying on-the-job skills. 
For example, many firms today still require applicants to complete coding problems on a whiteboard. In her blog, founder of TCB Analytics Tanya Cashorali stresses: “Whiteboard testing adds unnecessary stress to an environment that’s inherently high stress and not particularly relevant to real-world situations.” Also, candidates who might be top performers in real-world situations may not be the best test-takers, which is a lost opportunity for both the client and the candidate. 
To avoid the fallbacks of this popular interviewing method, take-home tests, real-life data science projects, hackathons, and online AI competitions should be utilized instead. These provide real-world challenges, evaluate actual technical skills, and provide realistic resource access as opposed to whiteboard interviews. With this approach, every candidate is given an equal chance to succeed. The result is getting the right fit for the job and optimal data utilization in your firm, so the hard work is worth it.



Tags

The Noonification banner

Subscribe to get your daily round-up of top tech stories!