5 Things To Look For When Hiring A Data Scientist
Features you should look for to hire the best match.
By Sham Mustafa •
Opinions expressed by Entrepreneur contributors are their own.
You're reading Entrepreneur India, an international franchise of Entrepreneur Media.
Critical insights, analysis, curiosity and intuitive skills are the hallmarks of a great data scientist, and the individual who possesses them can bring your firm many benefits. The role of a data scientist is becoming increasingly important to companies large and small, and it is a malleable one, at that. In a competitive field where demand for data scientists exceeds supply, this hire may be one that will have a long-lasting impact on your business.
The job of data scientist can vary from one organization to another. Some corporations seek applicants with great proficiency at data analysis and programming; others look for business and product intuition; still others look for both. There are also other features you should look for to hire the best match.
Your company will be investing a lot both in the interviewing process and the job offer. As the employer, you want to make sure you get it right. In a field of excellent candidates who have proven that they have the education, programming skills and backgrounds to succeed, how will you be able to identify the truly outstanding individual?
Here are some suggestions of what to look for when hiring a data scientist:
- Seek fluency in multiple technologies: databases, scientific computing, predictive modeling and data analysis. Test quantitative skills across various dimensions. You want someone who is actively seeking out the best tools for the job, for example, using Apache Spark to convert streaming big data and sensor information into actionable items.
- Identify a good communicator who can visualize insights. Ask your candidate to describe her work, approach, methodology and outcomes of several large projects at her former jobs. Can she tell a story using data? Will she be able to get her point across with both technical and non-technical audiences? Visualization is key to good storytelling with data. Great candidates are able to tell a story with data, without slipping into too much technical jargon.
- Pay attention to statistical skills. An applicant who has strong foundational statistical skills has a competitive edge. You need candidates who are good on theoretical fundamentals; can do questions on probability; and hypothesis testing; Bayes' rule; randomization; and Simpson's Paradox.
- Hire someone who demonstrates domain expertise in at least one quantitative area. Whether you are hiring in healthcare, higher ed, finance or another field, you want to hire an individual who can ramp up on the details of the industry. Ideally, you want an applicant who has demonstrated an aptitude for your industry, company, product and workflow. However, some top candidates possess a broad skillset which is transferable and can add value to your firm immediately. For example, candidates who possess advanced degrees in Physics are able to perform well in quantitative research roles at financial services firms.
- Look for a team player who has done exceptional past work. You want to hire a team player who has people skills, who will blend in with your existing data science team, and someone who is ready to adapt to any difficulties he encounters when problem-solving. Ask your candidates to walk you through a large, complicated project from its start to delivery. What do they think are the most important features of a good team player? Are they willing and ready to help support others when they meet obstacles? Data Science is a team sport and it is important to hire candidates who are collaborative in nature.
Some hiring managers in the field find that one of the hardest things to do in recruiting a data scientist is to find someone who has the right balance of scientific and engineering mentality. That is why many companies now include on-site testing as part of their interview process, to check the applicants' programming and data analysis skills. Depending on the industry, it is also very common for companies to give applicants take-home tests/projects that are domain-specific and would require candidates to spend several days to complete.
As an employer, prioritize skill testing over interview performance in order to refine your pool of candidates. Testing allows a scientific, modern way to analyze applicants. An important part of this equation is the ability to assess candidate success factors by testing and weighing an individual's hard and soft skills.
