The answer lives in this podcast

Answer extracted from the Digital Construction Podcast — listen to the full episode below.

🎧 Listen to the episode on Listenly

What is the most important skill for the next generation of engineers working with AI-generated solutions?

6 September 2026 Digital Construction Podcast · Devon Middleditch

← See the Digital Construction Podcast fiche

Critical thinking will be essential for the next generation of engineers because they will receive significantly more information generated by AI agents. If you cannot critically evaluate machine-generated results, you risk letting problems slip through — the key is to verify whether an answer actually makes sense rather than accepting it at face value.

As AI systems become more prevalent in engineering workflows, the role of human judgment shifts fundamentally. Engineers must develop the ability to question and validate outputs rather than treating them as authoritative. This is particularly critical when working with solutions that span multiple AI agents, each potentially introducing their own assumptions or errors.

The difference in mindset is stark. Previous generations of engineers relied heavily on first-principles thinking because they built solutions from the ground up, developing intuition about what results should look like. As Devon Middleditch explains in the Digital Construction Podcast episode, the next generation will face a different challenge: distinguishing between plausible AI output and genuinely correct output. When you receive a result from a machine, you need to pause and ask whether it actually "smells right" to you — and if something feels off, dig into the calculation logic rather than simply moving forward.

"If you're not a critical thinker, you're going to let problems through when given a result from the machine."

This shift reflects a broader change in engineering practice. Rather than solving problems entirely from scratch, engineers increasingly work with AI-assisted outputs, requiring a different skill set. A point detailed further in this podcast discussion is that developing critical thinking capability becomes an organizational priority — it's not something that emerges naturally if teams are trained only to interpret and implement AI suggestions.

The practical implication is immediate: validation must become a standard step in every workflow where AI output feeds into engineering decisions. This doesn't mean distrust; it means systematic verification. When an AI agent produces a result, a critical thinker asks: Does this align with my understanding of the problem? Have I checked the methodology? What assumptions is the model making, and are they valid for this context?

Middleditch's experience scaling teams across major infrastructure programs has shown him that this capability gap is already emerging. As discussed at length in the episode, organizations that invest early in building critical thinking culture — rather than simply deploying AI tools — will have a significant competitive advantage.

Building a culture of verification over automation

The transition is generational and cultural. Engineers who built their early careers solving first-principles problems developed strong intuition for "what right looks like." They naturally question outputs because they know the terrain. Newer engineers, trained in a world where AI assists at every step, may lack that intuitive baseline. The solution is deliberate instruction in critical evaluation, not just tool usage.

Organizations must teach engineers to evaluate reasoning, not just results. A correct answer produced through flawed logic can mask downstream problems. Conversely, an incorrect intermediate result might still lead to acceptable outcomes if the error is caught and corrected. The skill is knowing the difference — something only critical thinking develops.

See also