Why AI Startups' Fixation on 'Cracked' Engineers is Short-Sighted

Before we get to today’s column, OpenAI’s Chief Financial Officer Sarah Friar pulled back the curtain this weekend on how the ChatGPT maker’s consumption of computational resources and annualized revenue have grown at approximately the same pace over time. That isn’t super surprising, given that most companies would expect their revenue to grow with increasing levels of investment. (The real question is whether the company can turn a profit anytime soon!) Still, her disclosures give us a good benchmark for how much revenue growth we should expect to see with every new compute capacity deal OpenAI strikes in the future.

Onto today’s column…

As I wrote on Friday, startups are getting pickier about the software engineers they hire, as AI can increasingly perform tasks that used to require human programmers. These days, many startups say that they’re only interested in “cracked” engineers—a piece of Gen-Z slang for driven and technically gifted engineers.

No one could fault startups for seeking out top talent. Some have taken their fixation on cracked engineers too far though. 

For one thing, the term cracked has taken on associations with a tendency to be anti-social. It has become almost a satirical meme for startup founders to brag on social media about the awkward places their cracked engineers are willing to whip out their laptops—at parties, for instance. While hiring productivity-obsessed engineers might help your company write more code in the short-term, it won’t lead to a welcoming environment for future recruiting targets.