The best hiring practices are slow, personal, and deliberately hard to scale

Brian Chesky, chief executive of Airbnb, serves as co-hiring manager for the top 200 people at the company. He describes the conventional CEO approach, in which a chief executive hires only their executive team and lets those executives hire everyone else, as “fatal.” That framing is worth taking seriously. Chesky is not talking about a marginal process improvement. He is describing a structural commitment to remaining personally present at a scale most executives have long since delegated away.

The same logic appears at the level of calendar allocation. Jon McNeill, who served as president of Tesla, has said that 60 percent of his calendar consisted of interviews during his time there. He also reports that Elon Musk and JB Straubel would be the last interviews, a final check before anyone came through at a certain level. These are not policies designed for efficiency. They are policies designed to preserve signal, and they come at obvious cost to everything else on an executive’s schedule.

Marc Andreessen has remarked that technical product founders spending 40 percent of their time on recruiting is remarkable. The proportion is striking on its face. It suggests that the people building fastest are, in some cases, also the ones most willing to slow down around the question of who joins them.

I am the co-hiring manager for the top 200 people in the company. This is very radical. A lot of CEOs think it's their job to hire their executive and their executive team hires their team. I think that is fatal. Brian Chesky

Adam Ward, who has thought carefully about the mechanics of this kind of hiring, draws a distinction worth keeping. The most productive way to surface candidate referrals from existing employees, in his view, is not automated outreach. It is repeated one-on-one sessions, held every other week for six weeks, focused on going through names together. The contact itself is the point. On the interview side, Ward describes hiring managers meeting candidates three or four times before a final decision, a sharp departure from the single short screen that became standard in faster-moving environments.

The strongest argument for intensive process comes from what happens when it gets cut. Ward recounts what he heard from Michael Terrell about the consequences of removing work trials from a hiring process. The reasoning for removing them was straightforward: trials add load, and not every candidate wants to do them. The result, in Terrell’s account as Ward relays it, was that signal collapsed entirely. The company reinstated work trials. The episode illustrates something easy to forget when optimizing for speed: the friction is not a flaw in the system. It is the system. The discomfort of a real work trial, or a fourth conversation with a hiring manager, is where the information lives.

David Haber adds a related note from the evaluation side. Using too much reliance on AI coding tools is, in his view, a flag, suggesting a candidate may not be approaching the work in the right way. That observation fits with what the accounts above describe more broadly: hiring processes extended into real tasks and real trade-offs rather than abstract exercises designed to be completed quickly.

What connects these accounts is not nostalgia for slow hiring. It is a specific claim about where information comes from. A single interview, however well-designed, captures a narrow slice. A work trial, a referral conversation run across six weeks, a fourth meeting with a hiring manager, and a senior leader in the final seat all add different dimensions to the same question. The leaders describing these practices are not doing so because they have extra time. They are doing so because they have concluded that faster methods do not produce the same quality of answer. Whether the rest of the market catches up to that conclusion before turnover and bad hires force the issue remains an open question.