Hiring in the Age of AI: What to Interview For

December 1, 2025

Hiring in the Age of AI: What to Interview For

By Gene Kim and Steve Yegge

The following is an excerpt from the book Vibe Coding: Building Production-Grade Software With GenAI, Chat, Agents, and Beyond by Gene Kim and Steve Yegge.

In traditional software hiring, interviews typically focused on mastered languages, frameworks used, and memorized algorithms. However, as AI transforms coding, it also redefines what makes an engineering candidate stand out. As leaders in professional vibe coding, we now ask: What should we be interviewing for?

Here's what we recommend: In almost all cases, prioritize candidates who have already engaged with AI. If someone hasn't at least experimented with vibe coding, it's a potential red flag. Imagine interviewing a chef who has never tasted garlic or salt; you'd want to understand why. They might have a compelling reason, but you'd still question their curiosity about essential culinary elements.

We'd ask: Have they explored chat assistants and coding agents? Can they articulate what worked, what didn't, and why they are excited (or skeptical)? Their responses will reveal more about their mindset than dozens of checklist questions ever could.

We're not suggesting you only hire enthusiasts who write thousands of lines of AI-assisted code before breakfast. Rather, the goal is to identify engagement, interest, and curiosity. As you know, vibe coding elevates developers to higher levels of abstraction.

Communication skills, once a mere 'nice-to-have,' are now non-negotiable. In vibe coding, precise communication dictates productivity, outcomes, and ultimately, how frustrating your day might be. Candidates must effectively describe problems, provide clear context, offer actionable feedback, and direct AI assistants toward solutions without costly misunderstandings.

Another vital skill is the ability to read and review code at scale. With vibe coding, you might write thousands of lines of code daily—as Steve has done—but did you realize this often requires reading and understanding nearly ten times that amount? It's akin to reviewing the entire source code of a medium-sized open-source project every day. Through this process, Steve has identified many subtle issues, including a persistent rogue code deletion that AI repeatedly attempted. Many of the problems he encountered could have been caught with more careful attention.

We highly recommend conducting practical assessments involving AI interaction. Invite candidates to solve problems using AI coding assistants. This isn't 'cheating'; it's precisely how they'll perform their job. We would even go so far as to interview them on coding assistants and determine their proficiency with at least one. These tools are rapidly becoming the new IDEs, and for now, serve as crucial adjuncts to existing IDEs.

Observe your candidates carefully: Are they thoughtful in framing prompts, adept at managing context, and savvy at debugging model misunderstandings? Or are they struggling or fumbling? Are they imprecise, carelessly accepting AI suggestions, or overly dependent on AI to do their thinking for them?

Seniority may matter less now, as vibe coding is new territory for everyone. Both junior and veteran engineers are navigating this learning curve together. What truly counts is their enthusiasm for climbing and their speed of learning.

Who knows where the next set of obvious practices will emerge? Kent Beck recently speculated, "There's going to be a generation of native [vibe] coders, and they're going to be much better than we are at using these tools." It would be a shame to pass on such talent because they don't fit a preconceived image of a 'good programmer.'

One final note: From Steve's experience, he recommends at least one in-person interview, and for assessing core coding skill, one 'air-gapped' interview with no AI assistance allowed. This practice helps avoid accidentally hiring candidates who cannot code at all without AI (a significant red flag by late 2025) and/or actual AIs interviewing for the job (an increasingly common problem).

We've witnessed firsthand how vibe coding transforms hiring priorities, so adjust your interview filters accordingly. Whether you're hiring for a single role or reshaping your organization around engineering with vibe coding, this approach offers a blueprint for assembling an organization filled with great new leaders.

For more insights on effective AI-assisted development, check out Kim and Yegge’s new book Vibe Coding and their podcast Vibe Coding with Steve and Gene on YouTube.

---

About The Authors

Gene Kim

Gene Kim has been studying high-performing technology organizations since 1999. He was the founder and CTO of Tripwire, Inc., an enterprise security software company, where he served for 13 years. His books have sold over 1 million copies—he is the WSJ bestselling author of Wiring the Winning Organization, The Unicorn Project, and co-author of The Phoenix Project, The DevOps Handbook, and the Shingo Publication Award-winning Accelerate. Since 2014, he has been the organizer of DevOps Enterprise Summit (now Enterprise Technology Leadership Summit), studying the technology transformations of large, complex organizations.

Steve Yegge

Steve Yegge is an American computer programmer and blogger known for writing about programming languages, productivity, and software culture for two decades. He has spent over thirty years in the industry, split evenly between development and leadership roles, including nineteen years combined at Google and Amazon. Steve has written over a million lines of production code in a dozen languages, helped build and launch many large production systems at major tech companies, led multiple teams of up to 150 people, and has spent much of his career relentlessly focused on enhancing developer speed and quality. He is currently an Engineer at Sourcegraph working on AI coding assistants.