To avoid the trap of hiring 'good enough' people, make the interview panel explicitly state which current employee the candidate surpasses. This forces a concrete comparison and ensures every new hire actively raises the company's overall talent level, preventing a slow, imperceptible decline in quality.

Leaders often feel they must have all the answers, which stifles team contribution. A better approach is to hire domain experts smarter than you, actively listen to their ideas, and empower them. This creates a culture where everyone learns and the entire company's performance rises.

The most promising hires are often high-agency individuals constrained by their current environment—'caged animals' who need to be unleashed. Look for candidates who could achieve significantly more if not for their team or organization's limitations. This is a powerful signal of untapped potential and resourcefulness.

The "attitude vs. aptitude" debate is flawed. Instead, hire the person with the smallest skill deficiency relative to the role's requirements. For a cashier, attitude is the harder skill to train. For an AI researcher, technical aptitude is. The key question is always: is it worth our resources to train this specific gap?

In rapidly evolving fields like AI, pre-existing experience can be a liability. The highest performers often possess high agency, energy, and learning speed, allowing them to adapt without needing to unlearn outdated habits.

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.