New Codio CEO Doug Hughes argues universities must move beyond theory and start teaching students how to work inside real AI systems before the workforce leaves them behind.
INTERVIEW | by Victor Rivero
As colleges and universities race to prepare students for an AI-driven workforce, the pressure to move beyond theory and into hands-on technical learning is accelerating fast. Institutions across the country are expanding applied AI instruction into programs ranging from business and finance to marketing, cybersecurity, and data analytics — reflecting a broader shift in how higher education defines career readiness.
At the center of that transition is Codio, a platform designed to give students practical experience working inside real development environments using cloud infrastructure, cybersecurity systems, and AI tools similar to those used in industry. The company recently named longtime edtech executive Doug Hughes as CEO as it expands beyond computer science into applied technical learning across disciplines. Hughes previously led OnlineMedEd through a successful exit and has held leadership roles at McGraw-Hill Education, Knewton, and Encoura.
In this conversation with EdTech Digest, Hughes discusses why he believes higher education has reached an inflection point, what universities still misunderstand about “AI-ready” talent, and why the future of learning may depend less on what students know than on what they can actually do.
You’ve led multiple edtech companies through transformation and exit — what did you see in those roles that made you believe higher ed is now at an inflection point?
The most incredible thing is happening at warp speed all around us — all jobs are becoming technical, all at once. Normally, higher education moves methodically. However, we’re now seeing an unprecedented race to redesign the entire curriculum at most institutions to incorporate applied technical and AI skills in virtually every domain.
‘…we’re now seeing an unprecedented race to redesign the entire curriculum at most institutions to incorporate applied technical and AI skills in virtually every domain.’
University of North Texas just informed us they are adding an AI in Geography course. That’s pretty far-reaching. Business schools everywhere are scrambling to incorporate applied AI education into their major courses of study (e.g., AI for Marketing, AI for Finance, etc.).
I love seeing this because our system of higher education is unparalleled in teaching skills at scale. And it comes at a moment when this very system has been struggling to stay as relevant as it once was. So this is really a win-win for everyone. Students need these skills; colleges and universities need to remain relevant; and corporations need employable grads.
It’s rare to see the stars align so perfectly. A quick look at headlines illustrates this point. Ohio State, University of Florida, Purdue, DeVry, and countless others are in the news proclaiming they’re in this race to meet the moment. I never even dreamed of such an inflection point, and I’m grateful to be a part of it.
You’ve said universities aren’t producing AI-ready talent — what’s the most uncomfortable truth leaders in higher ed aren’t saying out loud?
Colleges and universities are incredibly capable of teaching at scale. Further, faculty who’ve spent their entire careers in the classroom are incredible servants and difference-makers, and I’m grateful for them and what they do. However, the skills students need now are applied skills, using the same tools they’ll use on the job.
It’s tough for faculty to start teaching those skills and tools without help from companies like Codio producing courseware to facilitate teaching and learning of applied technical skills at scale. We owe it to students, faculty, and employers to help them in any way that we can, as urgently as we can. In short, they need help to make this transition.
When you say “AI-ready,” what capabilities are actually missing when graduates show up on day one — and where does that gap show up first?
It shows up immediately in execution. Can someone take a problem, use AI tools effectively, and produce something useful? Most can’t. They don’t know how to prompt, iterate, validate outputs, or work within real systems like ChatGPT, Gemini, or Claude.
‘They don’t know how to prompt, iterate, validate outputs, or work within real systems like ChatGPT, Gemini, or Claude.’
The gap isn’t awareness of AI. It’s the ability to apply it under real constraints. The real shift happens as the role changes with the use of AI. For example, entry-level employees using AI to produce work transition from roles as doers to roles as managers of work, process, and output.
Importantly, the judgment required to police for quality and accuracy exponentially increases. AI truly is an amazing technology, but the implications of its use reach far beyond just being “the easy button.” It’s actually harder and more complex to work with AI at scale, which represents a significant challenge too few talk about.
Where, specifically, does the current higher ed model break down when it tries to translate knowledge into real-world capability?
It breaks at the point of application. Students are assessed on what they know, not what they can do. Assignments are often disconnected from real environments, so there’s no pressure to navigate ambiguity, debug issues, or integrate tools the way you would on the job.
This becomes especially critical when large volumes of output are being produced at epic speed and scale.
Hands-on, applied learning isn’t new, so what has actually prevented it from scaling inside universities until now?
Two things: infrastructure and incentives. Historically, it’s been hard to give thousands of students access to real environments at scale. That’s changing.
But just as important, universities haven’t been incentivized to prioritize applied outcomes over content delivery. That’s starting to shift as employers and students demand clearer ROI.
To be clear and fair, faculty have always had an appreciation for real-world application of the content they’re teaching, as that is the “relevance” that makes it all click for students. But this moment is different — it’s not about understanding how things are applied in the real world, it’s about using your hands, your mind, your knowledge, and your tools to do real work. That’s a pretty significant shift.
As technical skills move into fields like business, marketing, and finance, what’s the first thing those programs are going to have to give up or rethink?
They’ll have to give up the idea that applied technical skills are someone else’s responsibility. You can’t treat them as electives or add-ons anymore.
Programs will need to integrate technical skill development directly into core coursework, which likely means less time on purely theoretical material and more time working on application of knowledge in real-world systems.
The great news is that foundational knowledge and hands-on application both serve each other well, which can create a virtuous cycle and improve both outcomes and learner satisfaction.
A lot of platforms claim to make students “job-ready” — what’s the clearest evidence you look for that someone actually is?
“Job-ready” gets thrown around a lot, but the real question is whether someone can operate in the real world. If you hand them a problem, can they navigate the tools, make decisions, and produce an outcome that works? That’s where the gap is today.
‘If you hand them a problem, can they navigate the tools, make decisions, and produce an outcome that works? That’s where the gap is today.’
I’ve personally hired thousands of people over the course of my career, and I’ve always looked for three core ingredients in A-level talent: drive, desire to win, and a great attitude. That’s no longer enough. Now, leaders must look for strong technical aptitude and an exceptional attention to detail in virtually all domain areas.
The thing that’s both amazing and terrifying about AI tools is that it empowers a single human being to be exponentially more productive. That means moving faster than ever before imagined. The faster you move, the more ground you cover. The more ground you cover, the greater the margin for mistakes.
This requires not only the technical aptitude to work in these new tools, but also the judgment required to police the output and ensure quality.
Looking ahead, what early signals will tell us which institutions are truly adapting — and which are just rebranding the old model?
The real signal is what students are actually doing, not what schools say they’re teaching. Are students working in the same tools used in industry? Are they being evaluated on outputs, not just exams? Are programs integrating AI and technical workflows across disciplines rather than isolating them in one department?
If those things aren’t happening, it’s rebranding.
—
Victor Rivero is the Editor-in-Chief of EdTech Digest and Executive Producer of EdTech Digest’s Future Focus Forums (F3), bringing together edtech companies with top U.S. school district technology leaders for the most effective way to gather feedback on new and existing products. Write to: victor@edtechdigest.com
0 Comments