Opinion 3 top mistakes hiring managers make with data scientists

Published
  • May 08 2017, 6:30am EDT

Making a company’s first data science hire or working to quickly scale a data science team is a challenging and often frustrating undertaking. With the ever increasing demand for data scientists, more and more companies are confronted with their lack of tools and personnel equipped to handle complicated data problems.

According to Gartner, poor data analysis costs the average company $13 million every year. The massive losses associated with the rise of big data means that it’s boom times for aspiring data scientists. But for hiring managers who desperately need to find brilliant candidates the hiring process has become a rocky road littered with misleading resumes, poor demonstrated ability and capable candidates who lack real world business acumen.

At The Data Incubator we work with hundreds of companies looking to train their workforce in modern data analytics or hire data scientists from our selective PhD fellowship. We understand that finding the right candidates with the right combination of computation, technical, and communication skills is hard for all but the most seasoned hiring managers. So we compiled the three most common mistakes we see hiring managers make.

Understand a True Data Scientists Before You Hire One

Data scientists aren’t just rebranded software engineers, however many hiring managers are often easily impressed by experienced engineers, and software develops who have rebranded as data scientists. While their coding prowess is worthy of praise, data science requires a completely different approach because big data problems are … well very big.

Big data breaks the mold of traditional computation- with big data, all the data cannot fit into RAM and the traditional business intelligence calculations would take years complete. As a result a robust understanding of parallelization is key for any talented data scientist, but even more so is an ability to effectively implement distributed statistical computations.

Along the same lines, an engineer who has found second life as a data scientist will struggle with solving statistical bugs. Trained to find programming bugs, an engineer or developer could write impeccable code, but if they failed to reweight training examples then their predictions will be off.

The difference there is fundamental of how engineers are trained compared to data scientists. Engineers can build simple discrete rules-based models with ease, but these models are ill-suited to derive the more subtle insights from continuous-valued data and are leaving money on the table.

Solid statistical chops are necessary to overcome these challenges to build the next generation of scalable predictive models. Furthermore talented data scientists aren’t just able to create models, or build machine learning applications, they’re able to use these powerful tools to glean insights, and tell stories from massive data sets.

Move Past Buzzwords for Hiring Data Scientists

Our brains rely on a host of associations as cognitive shortcuts for everyday life. Without them, we’d be terrible at making decisions. However, when hiring a data scientist it’s really important to move past the “big data” buzzwords that plague the industry and move towards interview questions and challenges that provide the candidate a chance to showcase their talents.

For example a human resources screener could be screening for “Neural Networks” yet miss a great candidate who lists with experience tuning “Keras” and “Tensorflow.” Candidates who gear their resumes towards buzzwords may also be focused on conveying their knowledge of popular terms, and have little real practical knowledge required for job success.

Résumés can provide a quick summary of a candidate’s qualifications. As a result, they’re a useful shorthand for HR professionals–who in most cases don’t have the same domain expertise as hiring managers–to quickly filter candidates.

In many cases, that works just fine. But the shortcomings are get more apparent the more quickly companies change and come to terms with the reality of unconscious bias. Ultimately, there’s no substitute for experience and judgment, and digging deeper means first getting over our reliance on buzzwords.

The Right Type of Data Scientist to Hire for Their Company

At the end of the day making any sort of hire is a culture fit and a data science hire is no different. Digital departments often benefit immensely from hiring data scientists who embrace the “fail fast” mentality at the core of many startups, and startup-esque digital economy companies. These departments are generally overflowing with data generated from mobile, tablet, laptop or desktop sources.

Mobile apps, e-commerce, wearables, and digital advertising are just a few technologies that fall into this category. When data is plentiful, analytics often benefits from the unreasonable effectiveness of data — the idea that as we are able to learn from more data, we are able to achieve increasingly accurate models. Doing so certainly requires a deep knowledge of statistics, but a strong computational background is needed even more, which means that a data scientist with a strong engineering background who can quickly build and learn in real time.

Non-digital departments on the other hand require a more conservative approach to data science, and the types of professionals best suited for these roles come from a strong statistical background. These non-digital departments are often found at banks and other commercial financial institutions where risk underwriting can be a long, drawn out, but data-driven process. Because of the delayed feedback, these data scientists have to be very careful to vet models up front.

In this case, pure software engineering will not be nearly as useful for a company’s survival. Instead, a strong statistics background can ensure the credit models withstand rigorous statistical scrutiny, internal audits, and government regulators. Failing fast here for a more traditional non-digital company could mean millions of dollars lost or heavy fines.

The Data Science Hiring Takeaway

The challenges surrounding big data and data science problems are evident in the difficulties teams and hiring managers face when trying to find exceptional candidates. However, if executive management (CIOs & CTOs), data science team leads, human resources and team members work together to identify their department specific needs based on company type, culture, and technical requirements than these common pitfalls can be avoided with relative ease.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access