As companies increasingly turn to AI and machine learning, a clearer picture of what it takes to succeed with real-world AI is beginning to take shape. Beyond the small circle of tech giants and early adopters, a different set of skills and approaches is emerging as must-haves for enterprise AI teams.
Not every organization can compete with the likes of Google and Facebook for top AI talent. And it’s not just data science PhDs that companies are looking for. To meet their business needs, CIOs assembling AI teams are looking for subject matter expertise, software engineering skills, and the ability to translate learning algorithms into actual business value. Here, advances in machine learning technologies are helping pave the way.
“We’re seeing a shift right now,” says Scott Likens, emerging technology leader at PricewaterhouseCoopers. “There’s a lot of maturity and commoditization in some of the well-used ML, and a lot of the big providers have algorithms and AI models available. You’re able to piece together what you need, so you’re looking for high-end software engineers to glue together these different algorithms.”
Instead of hiring high-level PhDs to create new models, companies now seek blended teams to get the right data, and choose the right models to get the right decisions, he says.
Here is a look at how several organizations are assembling AI teams to solve business issues — and how advances in AI technology are changing the baseline skills necessary for success.
The fundamental roles of a successful AI team
A balanced team for an AI project will include three key people, says AI expert Monte Zweben, CEO and co-founder at Splice Machine.
First, a data engineer who can take the information a company collects and turn it into data that’s ingestible by AI and ML systems.
Second, a data scientist with domain expertise who knows that, say, weather can affect delivery schedules, or that particular mechanical issues can affect maintenance schedules. The data scientist will also need to be able to test out different algorithms to see which ones perform best, and then adapt them if needed to get worthwhile predictions.
Finally, a software developer is required who can incorporate all this into actual applications.
“These are the kinds of skill sets that we are looking for,” Zweben says.
For many organizations, success with AI is more a factor of balance in these three key areas than it is the number of PhDs that have been hired.
The power of blended teams
Online marketing company Urban Airship provides a textbook example of the shifts under way in how successful organizations approach AI. When the company first began thinking about using artificial intelligence seven years ago, it hired a PhD.
“The first machine learning model we introduced was around influence,” says Mike Herrick, the company’s SVP of product and engineering. It’s easy to track whether a person clicks on a link in their email. But tracking whether they visit the site later, and through some other channel, is a lot harder, he says, and that’s where the machine learning came in.
Urban Airship’s marketing platform touches more than a billion users on behalf of its enterprise customers, so there has been a lot of data available to work with. Over time, the company has expanded its AI team to three data science PhDs — and 15 engineers who put the data science to work.
Even the data scientists the company hires aren’t theoreticians, Herrick says. “They’re hands-on. They’re not pure academic theory people. If you just have the theory, you’ll spend a lot of money and get nowhere.”
That’s because Urban Airship leverages existing machine language techniques, open source libraries, and cloud services. “We’re not generally inventing new low-level technology,” he says. “We have no interest in pure theory. We’re just not geared that way.”
The skills required for the company’s AI projects include not just data science skills, but also product management, user interface design, software engineering, and product marketing, he says. “AI and ML really does take a cross-functional team to deliver on this type of technology.”
The most recent major AI project involved figuring out the optimal time to send marketing messages — not optimal in a general sense of “Monday mornings are most effective” but optimal in the sense of “John Smith prefers to look at these kinds of messages on Thursday afternoons.”
The project involved talking with Urban Airship’s customers to find out what they needed. As a result of these interviews, the AI team wound up going in a radically different direction than their original idea, he says — but it paid off.
Some beta customers have seen response rates double as a result, says Herrick. Customers also liked the fact that the emails were now spread out in time.
“When they were sending a notification on a special offer all at once, it would result in their systems getting inundated with users,” says Herrick.
Read the source article in CIO.