Hire for Give-a-Damn

The signal is care, not skill

Alexandr Wang published the rule as a blog post and titled it "Hire People Who Give A F***." The observable version he describes is behavioral rather than credentialed: "You can tell people who are just sort of like phone it in versus people who hang on to their work as like, it's so incredibly monumental and forceful and important to them that they do great work. It sort of like eats at them when they don't do great work. And when they do great work, they're so satisfied with themselves."¹ The claim is that the most reliable predictor of whether someone will succeed, and whether working with them is any good, is the magnitude of care they bring rather than skill, experience, or pedigree. In his account this is visible in interviews and interactions but not derivable from a resume; it is a behavioral and emotional signal about a person's relationship to their own work.

Wang ties the signal to Scale's business specifically. When the product is data quality, the gap between acceptable and excellent output is a function of how much people care about the details, so the people dimension is the business. More broadly, he argues that where quality is core to the model, no process substitutes for personally invested people.

Personal review as the operational form

Wang reports that he still personally reviews every hire at Scale, approving or rejecting each one, on the logic that if care magnitude is the primary signal, its detection cannot be fully delegated.¹ Adam Foroughi runs a structurally similar practice at AppLovin, where every new hire needs his approval, though he describes arriving at it from a different motivation: after taking over HR he found that auto-backfilling every departure meant headcount could only rise, so he forced managers to justify each role, and hiring fell from hundreds a year to tens.² Wang gatekeeps to detect care and Foroughi to suppress bloat, and the convergence on refusing to delegate the hiring decision is the point of interest.

The first-hires corollary

Brendan Foody at Mercor states a related but distinct rule about stakes rather than signal: for the first roughly ten hires, hold a maximum bar with no compromise, because those people shape the next hundred.³ The two combine cleanly, with give-a-damn as the trait to screen for and the earliest phase as the moment to apply it most strictly. Foody adds a complementary behavioral tell he looks for: "immense conviction in whatever idea they're telling me, but when presented with counter data, they change their mind very quickly," a reasoning-quality complement to Wang's care-magnitude signal.³

Adjacent framings and limits

The principle sits alongside other talent lenses. Hire for spikes values extraordinary depth over breadth, and high-temperature people accepts noise from high-variance performers to capture their occasional brilliance; both are compatible with screening for care as the underlying trait. Wang notes the honest limits himself: intense care and anxious neuroticism can look alike from outside, and care is easy to perform in a single interview even if his "it eats at them when they don't do great work" formulation is harder to sustain as a performance over time.