When hiring, top firms like McKinsey value a candidate's ability to articulate a deliberate, logical problem-solving process as much as their past successes. Having a structured method shows you can reliably tackle novel challenges, whereas simply pointing to past wins might suggest luck or context-specific success.

To hire for traits over background, Mark Kosaglo suggests testing for coachability directly. Run a skill-based roleplay (e.g., discovery), provide specific feedback, and then run the exact same roleplay again. The key is to see if the candidate can actually implement the coaching, not just if they are open to receiving it.

The purpose of quirky interview questions has evolved. Beyond just assessing personality, questions about non-work achievements or hypothetical scenarios are now used to jolt candidates out of scripted answers and expose those relying on mid-interview AI prompts for assistance.

To gauge an expert's (human or AI) true depth, go beyond recall-based questions. Pose a complex problem with multiple constraints, like a skeptical audience, high anxiety, and a tight deadline. A genuine expert will synthesize concepts and address all layers of the problem, whereas a novice will give generic advice.

For high-level leadership roles, skip hypothetical case studies. Instead, present candidates with your company's actual, current problems. The worst-case scenario is free, high-quality consulting. The best case is finding someone who can not only devise a solution but also implement it, making the interview process far more valuable.

Ineffective interviews try to catch candidates failing. A better approach models a collaborative rally: see how they handle challenging questions and if they can return the ball effectively. The goal is to simulate real-world problem-solving, not just grill them under pressure.

Since AI assistants make it easy for candidates to complete take-home coding exercises, simply evaluating the final product is no longer an effective screening method. The new best practice is to require candidates to build with AI and then explain their thought process, revealing their true engineering and problem-solving skills.

For cutting-edge AI problems, innate curiosity and learning speed ("velocity") are more important than existing domain knowledge. Echoing Karpathy, a candidate with a track record of diving deep into complex topics, regardless of field, will outperform a skilled but less-driven specialist.

The story of interviewing 600 developers to find one CTO highlights a key lesson: high-volume interviewing isn't just about finding one person. It's about developing pattern recognition. By speaking with dozens of candidates for a single role, you rapidly tune your ability to distinguish between mediocre and exceptional talent.