How to Fill Your AI Talent Gap

Artificial intelligence is changing everything—fundamentally reconfiguring industries, professions, and lives—and enjoying high-profile success stories with everything from autonomous vehicles to consumer electronics. Increasingly ingrained in the business world, AI represents profound opportunity for enterprises. Opportunity they are increasingly interested in taking: according to the Gartner 2019 CIO Survey, the number of enterprises implementing AI grew 270 percent in the past four years and tripled in the past year.

Unfortunately, finding the talent to implement strategic AI initiatives continues to be extremely competitive and will likely continue to be in the foreseeable future. To demystify the process, I thought we could look deeper into the skills AI practitioners need and what an ideal candidate might look like.

What skills do AI practitioners need?

One could argue that the skills needed for AI are similar to data science—math, computer science and domain expertise. However, the truth is that AI models are predicated on two things – automation and data. Increasing sophistication in automating key aspects of building, training and deploying AI models like model selection, feature representation, hyper parameter tuning, means skills need to be focused on model lifecycle and model risk management principles. These principles ensure model trust, transparency, safety and stability.

Data SMEs are increasingly important to understand how to acquire and leverage enterprise, commercial and open-source data across a wide array of datatypes. Data engineers help prepare and enable data pre-processing for building pipelines to train AI models at scale. For example, organizations are learning how to use their camera data that have traditionally been aligned to their safety and security teams but now are being considered to help drive improved customer experience. Accessing that data across a hybrid environment while still optimizing for the edge needs diverse skills around IOT, network, deep learning, policy and ethics.

Are there certain career backgrounds that are better for AI?

While neural networks take us one step closer to emulating the cognitive processes of the brain, we still have a long way to go. Backgrounds of experts in neural science offer a complementary perspective to the traditional blended skill sets of statistics, mathematics, software engineering and domain expertise. As we build enterprise intelligence that learns over time, concepts of neural science such as perception and cognition become increasingly important to scale our capabilities. For example, the rise of system architectures that provide "context" to individual AI models, by arranging these models hierarchically and/or connecting them at different temporal or spatial scales. This stems from a recognition that humans do not operate with very narrow tasks in mind. Instead, we connect these tasks across space and time to provide more contextual understanding.

Additionally, experts that are familiar with breakthroughs in certain industries are valuable to applying those concepts into new industries. For example, the consolidation of the retail and healthcare industries are unmasking new opportunities.  These opportunities have similarities to the challenges that banks have traditionally faced because they serve customers through their retail brick and mortar branches that are now more digitally fluent.
 

How can people obtain the skill set and training necessary to succeed in AI?

A growing number of people are augmenting their existing skill sets by acquiring deep learning knowledge through resources like massive open online courses (MOOCs) and Kaggle, a platform designed for data science competitions. Unfortunately, it is still rare to find people who can do it in real world scenarios. The classroom training is absolutely worth exploring, but it is no substitute for experience in the field.

It’s critical for organizations to deploy and rotate their AI teams across various departments to gain exposure and understand challenges that the business faces. This kind of experiential knowledge can be brought together in a center of excellence as well as carry experiences from across enterprises to create a better solution.

Teradata and AI

As the demand for AI skills continues to grow, it will be critical for businesses to not only find the right talent but to partner with experienced companies who can help them create successful deep learning solutions.
Teradata has successfully implemented AI across multiple industries, proving the technology as well as producing material business outcomes. We continue to channel IP from successful, field-based AI client engagements into accelerators that lead to faster time to value and reduce the risk of custom AI initiatives. Additionally, we use our client understanding to better inform our product strategy and roadmap to ensure we are incorporating emerging approaches, techniques and technologies for AI & deep learning into our broader product portfolio.

If you have an AI challenge and are looking for assistance, please learn more about our AI offerings.