Calling all rock stars: hire the right data scientist talent for your business


This is a contributed article by Scott Zoldi, Chief Analytics Officer, FICO.

Try googling “rock star” and “data scientist,” and prepare to be amused. It’s actually a thing…using rock star and data scientist in the same sentence. Don’t get me wrong, I get it. As a data scientist myself, working with some of the most brilliant minds in the industry, I’m amazed by the creativity, intelligence, vision and raw talent of my colleagues. They collaborate every day, harmonising their strengths and expertise around responsible AI to solve the big issues facing our business, our industry and our world. They’re working to correct financial inequity and disparity. They’re developing machine learning to stop financial crime and money laundering. They’re developing tools and platforms for others to leverage at scale. Their set list is long, and I’m proud to be their biggest fan and collaborator.

Executives sometimes say to me, “You make AI sound easy. How can my company get started?” First, it’s not easy, often complicated by the team structure and organisational philosophies at play. Second, you start by building a rock star analytics team — a carefully selected ensemble that balances each data scientists’ strengths, while also recognising and addressing capability gaps on the overall team.

It’s an up-front investment that won’t come cheap. Demand for data science talent is high. However, demand for AI products is also up since the onset of COVID-19, according to a recent Corinium survey. If you’re thinking of building your own group of analytic artists, here are a few guidelines to consider.

Set the stage: scoping out needs, capabilities

Before assembling a team that makes beautiful music together, you first need to stop, take a hard look at your organisation and ask questions. What are you trying to accomplish with this team? What resources do you already have in place — technology, expertise and executive sponsorship — to support this team? What are your company’s data analytic strengths and weaknesses, and how can this team impact those areas? How will this team engage and communicate with others within the organisation and deliver value to the business? Will this team engage externally, with customers and industry peers? What is your budget? How will you measure the ROI of the team?

There’s no template or magic formula for getting it right. In fact, 65 percent of AI leaders admit that building a team with the right skills is a significant barrier to AI adoption, according to a July 2020 Corinium report. Furthermore, it’s worth exploring how to incorporate greater gender and ethnic diversity as you set out to build your analytics dream team. According to a McKinsey report, companies and teams with greater gender, ethnic and cultural diversity outperform industry peers by up 33 percent.

It’s an iterative process where you ask the hard questions early and often to produce a successful outcome. First and foremost, the team should appropriately balance the company’s current level of analytics sophistication and aspirations for AI adoption. From there, you can determine the right size and capabilities of the team based on organisation-specific needs and objectives. 

Hire the right talent: identifying key positions

Once you set the stage, then you can focus on talent. The key here is diversity — look for a mix of skills sets and talents. Think of it this way: you only need one Elvis. In turn, he needs a band of great musicians to be successful. Indulge me as I run with this analogy and share my thoughts on key positions that comprise a rock star analytics team.

  • Project Manager. Think of this role as a drummer or bass guitarist. Coordinating all aspects of an AI project from start to finish, this key player “keeps the beat,” working closely with all internal and external partners and stakeholders to advance the project, set/manage expectations and ensure it delivers value to the business. Without a project manager to keep things moving, projects can quickly fester or fail, wasting precious dollars and resources with no final product.
  • Algorithm Developer. Machine learning is the algorithm developer’s “jam.” They have a deep understanding of a wide variety of machine learning approaches to a solving a problem, including the research work of others within the industry. The algorithm developer uses their rich knowledge base to create unique machine learning algorithms specific for the business problems of an organisation. Without this vital role, fundamental mistakes could be made before the first line of code is written, since their job is to ensure that the algorithms used to solve the business problem are sound and appropriate.
  • Analytic Software Engineer. This indispensable backstage crew ensures everything goes off without a hitch. They focus on deployment of the machine learning model, understanding the operational environment, enterprise software stack or cloud capabilities and providing core software. Analytic software engineers establish constraints for the data scientists to operate within so the end model can be deployed and meet business outcomes. Further, they also provide automated testing and essential monitoring of models. Without a constant focus on how the machine learning model will be deployed and monitored, many models will, ultimately, not be used.
  • Data Specialists. Just as lyrics make songs, data fuels models, making these experts invaluable. They ensure the data used to develop the model are of high quality and sufficient quantity, and they monitor it continuously. They provide insights to patterns and sensitivity of key data and relationships that can shift over time. Without high quality data, the initial model won’t be developed appropriately. Without their insight, key data relationships are missed. And last, without their pre- and post- production monitoring, the model becomes out of tune and, worse yet, organisations do not use AI responsibly.
  • Core Analytics Expert. As visionaries who bring “the art of what’s possible” to the stage, core analytics experts use their mastery of inputs coming from data, business need, software/architectural constraints and algorithms to design the appropriate machine learning model. Without their ability to seamlessly join the pieces, design the solution, lead the team and operate within the given constraints, analytic projects can be impossible to execute and operationalise.
  • AI Evangelist. This is your Elvis. AI and machine learning are hard concepts, yet these scientists can simplify and communicate complex data science solutions expertly for each audience, be it hardcore analytic skeptics, internal stakeholders, customers or partners. Without these customer-facing experts, machine learning and AI become science-fiction concepts which limits adoption and, consequently, recognition of the technology’s benefits.

In my (admittedly biased) opinion, today’s data scientists have earned their rock star status. They’re transforming our world with AI-driven processes that fuel next-level performance and better business outcomes. But, before jumping on the bandwagon, take the time to consider what’s right for your organisation. To build a balanced, functional team that fits the needs of your organisation, be selective when choosing your team and take the time to understand the unique role each scientist plays in the band.

Scott Zoldi is Chief Analytics Officer at FICO, driving the company’s innovation in artificial intelligence and incorporating it into FICO solutions. While at FICO, Zoldi has been responsible for authoring 110 analytic patents with 56 patents granted and 54 in process. He is an industry leader in developing practical applications and standards for AI, Explainable AI, Ethical AI and Responsible AI, and was named one of Corinium’s 2020 Global Top 100 Innovators in Data & Analytics. 

Read More …


Write a comment