Universities Are Facing an AI Reckoning

Welcome to this installment of Veo, where every week, I share a bit about what myself and others are seeing on the cutting edge of software product development.

In this issue…

Did someone forward this to you? If you like it, you can sign up here.

Want to grab time on my calendar? You can do that here.

-John Cox

I sat in a meeting yesterday with a committee advising an academic department at UT that was supposed to be about fundraising and alumni engagement.

We ended up talking about AI instead.

One of the attendees wanted to know how students should be learning to code. Should they still take Java classes? Should they manually type out lines of Python?

I told him I thought that was more or less a waste of time.

Not because coding is going away. Because the skill that matters now is completely different.

Working on projects while we’re developing VEO, I’ve learned more deeply about development - whether it be React, Typescript, or AWS deployment infrastructure - working with Cline and Claude than I could have possibly done on my own. Moreover I’ve been learning at a level that’s deep enough to really understand what’s going on yet broad enough to give me a big picture and to let me be able to check the agents.

I'd describe what I needed. The agent would build it. I'd review the output, spot the issues, give feedback, and iterate. Over and over.

The real work isn’t production - what is there to gain from me flailing away at the keyboard in a foreign language?. The real work is guidance.

Knowing what to ask for. Recognizing when something was off. Understanding enough about the frameworks to evaluate the output and then course-correct.

It's the same skill set you need to manage a team of engineers. Except the team never takes vacation, never forgets context, and works around the clock.

There was push back. How will students learn the fundamentals without writing code themselves?

Fair point. But you can learn fundamentals through this process too. You just learn them differently. By seeing patterns emerge across dozens of iterations instead of struggling through Stack Overflow for hours. Even better - train your agents to help you recognize and memorialize the patterns.

The conversation kept circling back to the same question: How do we prepare students for a world that's changing this fast? And how do we keep the faculty up-to-speed and embracing the change?

And then it hit me.

Universities aren't the only ones asking this question.

Every company with an engineering team is facing the same reckoning. Your developers probably already use AI coding assistants. But are they learning to manage AI agents? To orchestrate multiple specialized agents working in parallel?

Are your product people learning to structure work for agents instead of humans? To plan sprints around ideal ticket order instead of managing around headcount constraints?

The skills that got your team here won't get you there.

The gap between companies that figure this out and companies that don't is going to be exponential. One team will be moving at high speed and the other will still be standing still.

And the talent that understands this difference? They're going to gravitate toward the teams that get it. This is true for colleges and universities as well. AI-empowered students coming out of high school are going to gravitate to schools and programs running at their speed or better, or else they’re going to feel like they’re standing still.

I'm curious what you're seeing in your organization. Are people treating AI like a better search engine? Or are they fundamentally rethinking how work gets done?

Hit reply and let me know. I'd love to hear what's working and what's not.

-John

Three things worth reading this week: