In this installment of the IT Leaders Fast-5 — InformationWeek’s column for IT professionals to gain peer insights — Dr. Marc Hoit, CIO and vice chancellor for IT at North Carolina State University, talks about what it takes to scale technology across a large public university.
Since taking on the role in 2008, Hoit has led the development of a campuswide IT strategic plan and governance structure, guided by a longstanding belief that access, transparency and participation are essential to making technology work at the university level.
That mindset informs the initiatives he describes here, including the creation of a research computing and data unit designed to support faculty across academic disciplines as demand grows for advanced computing tools, AI and other emerging technologies.
Hoit, who is also a professor in NC State’s Department of Civil, Construction and Environmental Engineering, has reinforced that philosophy through an internship program that pulls students from across the university — helping to build buy-in and foster collaboration.
Communication is essential to technology adoption, especially in cybersecurity. He pointed to a recent security protocol change that, despite advance notice, still caught some users off guard.
This column has been edited for clarity and space.
The Decision That Mattered
What recent decision — technical or organizational — has made the biggest difference, and why?
The biggest decision that I’ve made recently was creating a research computing and data unit within my organization. It provides consulting-level and technology implementation support to faculty to meet their scholarship and research needs.
We used to have a high-performance computing [group], the server groups and all these other groups. When we started with those services, they were used by people like me who knew how to write the language, debug it and set it up. We were all fairly independent — just give me an account and a computer.
Now fast forward. With the current demand for research computing, GPUs and AI, the way [people use] technology has shifted. It could now be anyone from the humanities to the business school, to agriculture and life sciences to veterinarians. Everyone wants these AI tools, but most don’t have deep [backgrounds] in computation.
We’ve been building the research computing and data unit over the past five years. The faculty needed more support, so we partnered with the NC State University Library, and we’ve dedicated a small number of FTEs [full-time equivalents] to [act as facilitators]. They reach out to faculty, find out what their needs are and document how to address them.
Now we’re getting to the point of helping faculty to architect and design [systems] — for example, designing a database that they can run on their desktop.
The Hard-Won Lesson
What didn’t go as planned recently — and what did it force you to rethink?
The struggle is always around implementation, communication and partnering. We instituted some security controls because there have been a lot more cybersecurity incidents. We’re not immune to it. We do our best to stop them, and for the most part, we do.
But security controls are always hard, because if you get phished, we lock your account because we don’t know what you’ve done. Sometimes, [the threat] can be so deeply embedded, no matter what you do to that computer, when we bring it back online, it just regains access [to the network] and causes problems.
A few things we’ve done have overwhelmed our help desk. The most recent one is we turned off SMS for two-factor/multifactor authentication. When you get a text message with the code number that you put into the application, that’s actually considered risky by NIST, the National Institute of Standards and Technology, which provides security rules and policies. They have said that SMS is not a secure method for second-factor authentication — it’s too easy to spoof. Our team decided we should turn it off, and we have another protocol that we use.
We communicated and tried to tell [the university]. Monday morning after we turned it on, there were 2,000 people waiting in line and getting frustrated with our help desk. We couldn’t answer their calls fast enough and open their accounts back up.
The challenge is about communication — we do the planning, we think we’ve communicated, but they didn’t read their email. You don’t want to put announcements in a newspaper — when you talk about security you don’t want it to be public because then the hackers know about it.
There’s this balance of how much to communicate, how to get the buy-in that you need. And inevitably, there’s a surge [of requests], because people tend to be less responsive about doing things in advance.
We keep trying to improve our communications and find other ways to get people involved. We ramp it up to try and head it off and we believe we’re getting better, but there’s always something that we didn’t think about.
The Talent Trade-Off
Where are you investing in talent right now — and what are you consciously not investing in?
The entire UNC system is under some heavy restrictions on head count.
I’m a faculty member, and I’m used to working with students a lot, so I instituted an internship program. We hire 25 interns a year and pay them a decent wage. We put them in our environment, and within about 8 weeks, they’re helpful for that semester. They’re not competing with staff, but they’re doing many of the same things a staff member that’s been there for a while could do, certainly a new staff member.
We source from across the entire university based on the area we need to work in. I have a communications department, so we hire from the communications group. [Interns] from the management school are focused on helping with the business side of technology as opposed to being pure technologists — so they’re a good source. When we need hardcore coders and cybersecurity people, we go to computer science.
It’s part of our accreditation — demonstrating that we’re doing “high-impact teaching.” The students don’t just sit in the classroom and take tests, but we have them doing senior design projects, working as interns, and these different avenues give them a much more impactful education.
If I can keep interns for four semesters of outstanding work — bringing innovative ideas and new talent — some of them I can hire when it comes time to graduate. They get the benefit of working in the real environment. When they go out looking for a job, the people interviewing them say, “You have more experience than most of the people I’ve interviewed.”
I’ve also worked really hard on the reputation of the NC State University OIT [Office of Information Technology]. At the high-research universities like ours, we have self-organized meetings twice a year. There are about 50 of us, and we share things that, in the business world, would not be shared. The group includes CIOs at Research 1 universities, specifically the AAU Group (Association of American Universities). We don’t share names and anything secure, but we share techniques and we learn from each other. In building that reputation [within that elite group], when [peers] see an opening here at NC State, it’s very attractive — and that helps us a lot in hiring.
The External Signal
What recent external development is most likely to change how your organization operates, even indirectly?
AI is kind of in that chaos and overhyped stage. We are obviously looking at it — our faculty, our staff, our students — they’re all looking at it and using it for personal productivity. My units are looking at how we can get it to an enterprise-class usage, which is a little harder right now. We’re also a Google school, and Google is making great strides in the tools we use that are automatically built in.
The other external development impacting us are the budget shifts — and the frustration at the federal, state and UNC level. There’s no one group that’s causing the problem — [there’s a mandate] to cut research overhead, for example, amid uncertainty about the budget. We still don’t have a state budget for the university. What that means is that some of the new dollars we were expecting to use to help fill some of these extra positions and to grow the research effort that we’re doing to support the faculty is at a stalemate.
The Perspective Shift
What have you read, watched, or listened to recently that changed how you think about leadership or technology — even slightly?
May I change it to [encompass] readings and work that I’ve done over my career? I’ve found that transparency and inclusion of people — meaning your team members, your direct reports — sharing with them the ideas and bringing them into the conversation, including colleagues and everybody else, creates less difficulty when you want to implement something that they’re part of.
Collaboration is the outcome, because if we’re all planning, designing and doing it, then you’re going to be mostly happy with the result. If you said, “We’ve had a group of three people, and we designed this for you — here, use it,” there tends to be a more negative reaction.
Enjoyed this article? Sign up for our newsletter to receive regular insights and stay connected.