An interview with Joe Inzerillo, EVP and CTO, MLBAM

Part 12 in a series of CTO Interviews

Recently I had the opportunity to sit down with CTOs from very different digital media companies to understand how they think about the role, what they have in common and where they differ in approach. You can read earlier interviews with the CTOs of companies such as AOL, The New York Times, Mashable, Buzzfeed, Vice, Salon, Shutterstock, Digg, News Corp and Hearst here.

Next up is Joe Inzerillo, EVP & CTO of MLB Advanced Media. MLBAM was recently described by Forbes as “The Biggest Media Company You’ve Never Heard Of.” According to that Forbes report the spinoff of Major League Baseball has built a nearly $1 billion technology business around the capabilities it perfected initially to deliver live and on-demand baseball content on MLB.com. Now its clientele includes many major media companies, from ESPN to, reportedly, HBO. MLBAM is an NYC Media Lab member company- currently the Lab is hosting a challenge on automatic video annotation using archival video from across baseball history. I caught up with Joe in MLBAM’s headquarters in Chelsea Market.

I’ve started each of these interviews asking CTOs to address whether the pace of technological change has changed the way they think about their jobs. Are things really more fast paced these days?

I definitely think the rate of change is increasing. The reason for that is the accessibility of tools. Look at Amazon Web Services- before you had AWS, if you wanted to be an internet company you had to build servers, you had to put them in a data center; the barrier to entry was relatively high. Now you can sign up in a minute, and as long as you have a credit card you are off to the races. The industry has reduced some of the friction that it takes to get going. Therefore, more companies enter the funnel, and more companies pass through the funnel to build a product that you might actually want to look at.

Ten, fifteen, twenty years ago, you used to call your trusted vendors and you asked them, ‘hey, what’s new?’ They’d come in every quarter. Now, these are not necessarily the companies that are driving innovation- instead they tend to acquire the smaller companies that are in fact driving innovation. So if you want to be on the cutting edge, you have to take a lot of meetings, some of which are very productive and some of which are a complete waste of time. You are doing technology business development as a third of your job, in addition to your regular job, just to stay current.

How do you build an organization that can keep up with the pace of change?

I’ve empowered my lieutenants, and in turn their teams, to take a lot of this burden on individually in their specialties. Their work matriculates up, and if we decide we will action something, we get a larger group together to consider the proposition. These days my tech teams know more about evaluating a business than they did ten years ago; part of the evaluation of these young companies is whether they are viable.

Are you generally more external facing than you used to be?

We have always been a web company, so we’ve always had this discipline. But in the last five years, as we’ve expanded the business beyond baseball and have taken on so many partners, that ups the ante because we see a lot more use cases than just narrowly the baseball use case.

You have to create a culture that is not failure averse. That is something that is very difficult. You have to try a lot of things, but you have to do them quickly enough that you can identify failures quickly. If everything you do makes it to production- sees the light of day from a user standpoint- then you can’t possibly be innovating. Nobody bats a thousand.

With regard to innovation, in practice we depart from the very large tech companies- the ones that say, for instance, take one day out of the five to go and do self-directed research. We don’t mandate a specific amount of time that people should spend on innovative projects. Rather, I say we want people to do work that is still related to what the business is doing, but they are welcome to take chances, knowing they may fail. We’re trying to strike that balance- keeping people accountable for the workstreams that fuel the business and at the same time embracing experiments that might fail.

You seem to excel at combining technologies from different disciplines- I think of an example you once gave about combining radar technology designed to track missiles with software used to improve docking boats in a wharf. How do you get at that combinatorial approach?

Part of it is that the dynamic range of technologies I deal with is automatically large based on the scope of some of the problems I look at. Just consider baseball- I’m looking at technology on the television side, online, the back of house tech that runs player operations, plus all the stuff we do for third parties which spans from wrestling to other sports to standard entertainment and television content. So the range makes it a challenge, and changes the complexion of my staff that is helping me with those decisions.

Fundamentally I’m big subscriber to the philosophies of Benoit Mandelbrot. Mandelbrot had this idea about intellectual fortresses. This was the thought that you can’t understand a specific problem outside of your discipline. This is a computer science problem, or a hardware problem, or a legal problem, for instance. Mandelbrot believed that these intellectual fortresses stifled innovation. I believe that this happens today at a more granular level. Go back five or six years ago, you heard, oh well, that’s a job for a database administrator, or that’s a job for a network person. What you now see is that the millennial companies- no one ever told them “you can’t do that.” They don’t understand the difference between the roles. So when you look at DevOps as a movement, it shows that one person can be very empowered and productive to use the toolset to move something from point A to point Z.

That brings me to hybrid technologies. When you start to break down those barriers, there is no question that people with different backgrounds are going to come at a problem differently. The technology that emerges from that problem solving is inherently different and incredibly complementary. When you look at complex systems, they are almost always hybrids of different types of technology, as opposed to a singleton growth path where you plant a seed and get a tree. It’s much more like building a building- get some steel, some drywall, other elements, and you build a house. That’s more of the metaphor that we’re looking for. The successful things we have done have been a fusion of different perspectives, something like At Bat, which has a lot of video engineering, data analysis, user interface, mobile engineering- all wrapped into a product that kicks ass.

You’re known for doing a lot of work with universities. And of course, you’re working with us. What do you get out of your engagement with schools?

I think there is an obligation to society as a whole to promote education- part of the mandate of being from the sport of baseball is that we are part of the fabric of society. On the technological side, one of the things that I love about working with students- those in college or graduate school- is they aren’t afraid of failing. They come at a problem from a different perspective, no one has told them their approach is a bad idea. Sometimes it is the wrong approach, but sometimes it ends up being the most effective approach. So tapping into that vision from the next generation enhances us. It gives them a path into the organization and perhaps a win under their belt that maybe they can parlay into a career. So it’s a very symbiotic relationship where we can feed off of that energy and thirst for knowledge- unafraid, unabashed enthusiasm to try to attack a problem.

I think the challenge we’re running with NYC Media Lab goes back to the hybrid technology notion. When you think about what is machine vision, it’s still so nascent but we are so far from where we’ve come. When you approach a project like this challenge, there are a lot of techniques. If you just look at this as a machine vision problem, you might spend a lot of time tuning to read graphics, numbers of the players etc; but if I can find the roster of the people on the team, I can narrow the possibilities and build a better system from that restricted set of data. So that approach means taking other technologies, federating them in to synthesize a result. Those are the folks that will win this challenge. It is the type of nimble thinking we are trying to encourage, and the type of nimble thinking that is crucial to continue to advance.

What kinds of technology problems do you think will drive the next wave of innovation?

There are some problems that are incredibly difficult to attack. One of the hardest problems right now is the ability to come up with better and better mechanisms to separate business logic out from presentation. As we get into a much more fragmented world of mobile devices, internet of things equipment, apps- we still don’t have a great way of having a rich user interface without endowing that user interface with a whole bunch of business logic. There is still an opportunity for a watershed moment for a technology, a language, a process that solves it.

What we have right now is silos- the bright line between front end and back end. You really want to be able to do something smarter. My guess is that the way we’re going to do it is much more machine automation: not in abstraction or middleware, but where the machine will write the code. You look at some of the artificial intelligence languages of the past, languages like LISP, they weren’t very effective. But I think in the future that type of technology will continue to evolve the state of the art. So artificial intelligence will be a big part of our business. Computers helping people write code will help us address these challenges.

Justin Hendrix is Executive Director of NYC Media Lab. Reach him at justin [dot] hendrix [at] nycmedialab [dot] org or follow him on Twitter @justinhendrix.