Incorporating AI, Innovation To Meet Business Objectives, With Help from Amazon AWS

Kesha Williams is a software engineering manager in the Information Technology department at Chick-fil-A, the restaurant chain, with responsibility to lead and mentor junior software engineers and deliver new and innovative technology capabilities. She has worked in IT for over 20 years; she started at Chick-fil-A 13 years ago, working 

Q. Can you tell us about your role in information technology at Chick-fil-A?

A. I’ve been with Chick-fil-A for the last 13 years, and I’ve worn many hats during my time there. In my current role, I serve as a software engineering manager; that entails many things. First, I’m responsible for just leading and mentoring junior software engineers. I also help to define and implement the onboarding process for our new engineers. And I develop technical training curriculum to keep all of our engineers up to speed on the latest and greatest technologies. And then I’m also able to code and build solutions that use Angular, Java, Spring and Amazon Web Services.

And in a really fun part of my role, I lead innovation teams. We investigate emerging technologies like artificial intelligence, machine learning, computer vision, virtual and augmented reality, and voice-first technology. I lead teams investigating how we can use these technologies to help move the business forward. And then, lastly, I serve as what we call a technical evangelist for Chick-fil-A, so I speak at conferences on Chick-fil-A’s behalf and share some of the cool and exciting things we do in IT. So that’s my role. Yeah, it’s a lot. It’s a lot of fun, too.

Q. It sounds like a really interesting job. Where does AI fall on the strategic roadmap of Chick-fil-A? And how would you describe the degree of support for AI technology investigation from the top management at the company?

A. From an IT perspective, we exist to power the promise of the brand, so top management and IT work together to foster a culture of innovation. IT is responsible for delivering new technology capabilities to enable business success. We are encouraged to innovate and think outside of the box when we are trying to solve business problems. And of our top five projects for this year, three of them have an AI component, either directly or indirectly. So we definitely see AI as a technology that will help us deliver new capabilities to the business.

Q. Can you talk about what AI technologies you are exploring for the company and/or could you talk about any of the projects that incorporate AI that you’re working on?

A. I can speak at a high level about some of the AI technologies that we are exploring. We are exploring machine learning, computer vision and facial recognition, and voice-first virtual assistants. One of the innovation teams that I led won the Think Different Innovation Award from Chick-fil-A for thinking outside of the box in developing Amazon Alexa skills, for use by Operators in the restaurant and by customers outside of the restaurant.

And we’ve explored machine learning to assist with making reward and “treat” recommendations via our Chick-fil-A One Mobile app.

Q. Can you talk about what the virtual or augmented reality could be used for?

A. We have a Chick-fil-A One mobile app, and we’re always seeking ways to have customers engage with mobile and online ordering. We’ve discussed that augmented reality could be a way to gamify the mobile ordering app. So, for example, we have the Chick-fil-A Cow Calendars. Each month, there is a free treat that you can redeem at your local Chick-fil-A restaurant. But there are some months where we have mystery items, so a customer will not know what the free treat is until that month.

We’ve discussed that augmented reality could be a way to gamify the Chick-fil-A One mobile app so customers can hold their phone up over the calendar, and see the mystery item revealed on the screen. The customer might see a chicken sandwich or a milkshake pop up in place of the mystery item. It’s a fun way to engage customers to encourage ordering through the mobile app.

Q. Could you talk about how you got started in software engineering yourself?

A. I’ve been in IT for about 23 years, and I’ve been coding and building apps since high school. At the time, I didn’t realize it was called coding. I just called it playing with the computer. My father purchased a personal computer for our home when I was in high school, and he placed it in my playroom. So we had this room that was a playroom/office, and I just always joked with people and say I had a Barbie doll in one hand and the computer manual in the other hand writing code using the BASIC programming language. I’ve just always loved computers and technology. And because I was exposed to coding at such an early age, when it was time for me to decide what I was going to do with my life, and decide on a college major, it was easy for me to pick computer science and mathematics. Then after a summer internship with the NSA, I knew that I’d made the right decision. It’s been full speed ahead since then.

Q. You like to work with students, guiding them as interns at the start of their careers. What advice would you have for young people pursuing a career either in AI, if they’re interested in that, or IT, or what kind of exposure they need, what kind of experience? And, if they’re in school, what should they study?

A. I would definitely recommend having computer programming as a baseline. That’s what I learned when I was at Spelman College. I learned all the different concepts needed to build and write computer programs. It’s a solid foundation. That knowledge can be transferred to all the different areas in IT.  Specifically for students that want to pursue AI, I would recommend learning the Python programming language. I also see voice-first technology and virtual assistants as an easy entry point to AI because it’s so easy to get started, and you can quickly build and develop solutions that can be in the hands of millions of users in a very short period of time. So you receive instant gratification and feedback. That’s what I would recommend.

Q. Chick-fil-A is a big company, with many franchise owners. How does the pursuit of innovative technologies, including AI, fit within the whole organization?

A. In IT, we pursue innovation because it’s our job to deliver new technology capabilities that enable business success. So innovation is weaved into everything we do. For example, once a quarter we have an Innovation Day. This is where we are able to step outside of our day to day routine, our day to day job, and tinker with emerging technologies, mostly AI, to see how we can improve restaurant operations and customer experiences inside and outside of the restaurant. Through the Innovation Day and some of the teams that I’ve led, we’ve worked with Amazon Alexa, augmented reality, IoT, machine learning, computer vision and facial recognition. And one team that I led won the Think Different Innovation Award from Chick-fil-A for just thinking outside of the box.

Apart from that, we’ve recently opened a new Innovation Center in TechSquare, which is this huge technology hub in Midtown Atlanta. This is separate from our headquarters, and, there, we partner with universities like Georgia Tech to have year-round interns. We call the interns Innovators. We work with them to use emerging technologies and AI to solve business challenges. So we have a strong culture of innovation. I was lucky enough to be involved with the opening of the Innovation Center, and there are a lot of smart students out there.

Q. In your pursuit of AI and innovation, I am sure you work with many companies in the software industry. Are you finding is there anything that you need that is difficult for the industry to provide? Do they meet your requirements, especially for AI, or are there things that you could use that you’re not finding from the software industry?

A. Before the advent of cloud services like Amazon Web Services, I would’ve said, yes, there’s a lot lacking in the industry as far as support for AI. But now, thanks to AWS services like Amazon Machine Learning, Amazon Rekognition, which is their computer vision service, and Amazon Alexa Skills Kit, we’re now able to quickly build prototypes to help prove out the feasibility of our innovative ideas. And Amazon has also recently introduced SageMaker for advanced machine learning algorithms. Also there’s a new product called DeepLens, which I can’t wait to start playing around with, that allows for running deep learning models locally on a camera and to analyze and take actions on what it sees. There are so many opportunities to provide innovative solutions to the industry thanks to services like AWS.

Q. That  sounds like a big recent change. We have written about a talent gap between the AI expertise that companies need from the employees or new hires, versus what is available on the market at a cost that companies can afford. Are you able to find the IT staff you need to support your efforts?

A. At Chick-fil-A, we typically look within and train within to find this talent and AI expertise. So Chick-fil-A hires really smart people that are just overall great software engineers. They’re able to learn new things quickly. We all have a desire to innovate and a passion for technology, so we look within.

Q. So you avoid that whole issue of salary competition.

A. Right,we haven’t run into that.

Q. Do you have any concerns about issues relating to the increased use of AI technology, such as privacy concerns or data bias? Are there any that you have to be aware of?

A. Personally, I am definitely concerned with bias, data bias and machine bias. I speak a lot about this topic at technical conferences. I always go back to the classic examples like the Google AI program that classified a few African Americans as gorillas instead of humans, or the computer vision program that couldn’t even see African Americans. To me, that speaks to a lack of diversity in the data used to train the AI program. I believe that our reliance on AI, which is just going to continue to grow, makes a very strong case for the need for diversity in technology. We should build systems without bias.

In my spare time outside of work at Chick-fil-A, just because I wanted to learn more about machine learning, I created a predictive policing machine learning algorithm called SAM, which stands for Suspicious Activity Monitor. SAM looks at a particular situation using computer vision, and predicts the likelihood of crime using machine learning. SAM looks at several attributes about the person and their current location in order to make a crime prediction, so just think pre-crime from [2002 film] “Minority Report.”

When I created SAM, I intentionally excluded race as an attribute he considers because I didn’t want him accused of racial profiling. So, you know, another programmer may not have considered this, but just that decision to exclude race, for me, was an aha moment because it showed that machine learning can actually, if we use it correctly, remove human bias from certain situations.

When I switch gears and go back to Chick-fil-A and our use of AI, you know, in our industry, we pride ourselves on having outstanding customer service, which allows us to form emotional connections with our customers. If we start to rely too heavily on AI, that human element may diminish, which would really have an overall negative effect. So we just have to keep that in mind as we start to rely more on AI, that we always want to have that human element so that we can form strong emotional connections with our customers.

Q. Is there anything you would like to add or emphasize?

A. I would like to put in a plug for a recently-released AWS machine learning course I created with Manning Publications. We would like to offer AI Trends readers a discount.

(For AI Trends readers: Get a 30% discount on Kesha Williams course developed with Manning Publications, AWS Machine Learning in Motion.  Click on AI Trends Discount and use the discount code “aitrendsawsml.” The code is good (initially) through July 15, 2018.)