In Defense of SMEs

LinkedIn recently released its 2020 Emerging Jobs Report, and unsurprisingly titles like AI Specialist and Data Scientist are near the top of the list in terms of roles that are seeing significant increases in opportunities. It’s noted that this growth is due - in part, at least - to these roles assuming the work that other “legacy” roles have performed in the past.

There’s nothing wrong with that. Certainly there have always been shifts in industry trends and in “hot” skills that have resulted in the emergence of new standout employment roles, while others have faded. It’s part of the natural evolution of things. However, one of the things that’s not mentioned in the report, but which is an important consideration, is that even as data scientists and AI or machine learning specialists assume more work from other more traditional roles, there will remain a need for Subject Matter Experts to work hand-in-glove with these 21st-century dragonslayers.

The breadth and scope of what AI and ML can do is well-known and continues to expand. Increasingly, these technologies, when wielded by skilled professionals, can detect patterns, relationships and dependencies that were unknown to even the most learned experts in the field. However, the speed at which those sorts of insights can be gathered will continue to be improved by having true subject matter expertise alongside the AI or data science experts.

When building out models to do predictive analytics, there will be literally hundreds, or thousands, of possible features to select for inclusion. SMEs will be able to guide these model-makers with their own insights into what is known to be important, and what might be anecdotally suspected to be important by them and other experts in their field. Additionally, SMEs can help to describe meaningful constraints and limitations of the real-world problem that may not be obvious to the data scientist.

The ability to contextualize the mathematics and predictive models is one that for now remains an important role for humans to play, as artificial intelligence gains steam. Placing your team members in a position to understand how their jobs - and the data created by or consumed by their jobs - contribute to the application of these leading-edge technologies is one way to ensure that your organization gets more value - and thus more of an edge - out of the AI/ML and data science practitioners you hire.