skaman306/Getty Images
- Startups are scrapping with Big Tech over AI talent.Salaries and equity are peaking, as startups emphasize ownership and impact, recruiters said.The AI boom is fueling a newfound intensity in startup work culture.
Tech giants aren't the only ones fighting over top AI talent.
From multibillion-dollar acquihires to poaching packages reportedly worth tens of millions of dollars, the talent wars are having a "bit of a waterfall-like effect on comp" across early-stage startups, said Cristina Cordova, a former VC and the COO of issue tracking company Linear.
Salary and equity packages are climbing, and startups are touting the unique impact and ownership of working somewhere smaller, recruiters told Business Insider.
A head of AI at a Series A company — with an applied background and time spent in a top research lab — can command between $300,000 and $400,000 in base pay, said Shawn Thorne, managing director at executive search firm True Search.
That's without any previous management experience. Before the AI boom, the salary for a vice president of engineering may have been closer to $200,000 to $250,000, he said.
"There are two sets of rules," Thorne said. "The price that you need to pay for all talent, and then there's the price you need to pay for AI talent."
It's not just base pay that's exploding. Equity is "the big factor" because startups compete with the "opportunity cost" of a top researcher or applied engineer launching a startup of their own, he said.
Equity packages have climbed to between 2% to 5% for top engineers, Thorne said — compared to what would've been a fraction of a percent for an early-stage, non-managerial employee in the past.
Other bargaining chips include letting AI researchers or engineers come in with a cofounder title, compute access, and time for research outside work, Thorne said.
"We've seen top candidates win with remote flexibility, front-loaded equity, big sign-on bonuses, tighter cycles on compensation refreshes, and friends on the team," said Natan Fisher, an investor and the cofounder of recruiting service SingleSprout. "It's not just the offer. It's how fast and personally the process goes."
The fight for AI talent has become unprecedented. In June, Meta spent $14.3 billion on a Scale AI investment and what was seen as an acquihire of Scale AI's CEO, Alexandr Wang, while Google snapped up Windsurf's top talent this month for $2.4 billion.
OpenAI CEO Sam Altman previously said Meta had attempted to poach talent with $100 million signing bonuses, which Meta's CTO Andrew Bosworth said OpenAI had countered.
Startups can offer ownership and impact
Frenetic job-hopping has become the new normal in AI, Cordova said, contributing to "a mercenary mindset" and a higher bar for employee retention.
Startups have an edge. They can offer the opportunity to be part of something transformative and the potential financial windfall that comes with that. They can also offer a sense of freedom, as opposed to being a cog in a machine with a limited scope. (Big Tech CEOs like Meta CEO Mark Zuckerberg are well aware of these dynamics, similarly offering fewer reports and more AI compute power to woo top candidates.)
While Fisher said he doesn't know any early-stage startups that can compete with $10 million compensation packages, "they can win by pitching sharper problems, faster cycles, and giving top AI engineers the ability to own product, with strong upside."
The larger talent wars aren't affecting all startups, including those building off existing AI models. Shie Gabbai, the COO of AI travel planning company Layla, said that while nearly all of the company's funding has gone to hiring engineers, salaries have not been affected.
Gold star AI candidates
AI startups face a widening gap between gold star researchers — the kind selling their startups for billions — and rank-and-file employees who might not have the same pedigree.
The latter may fetch salaries more in line with what's typical in Silicon Valley, said Mark Bai, a managing director at True Search.
As AI increasingly replaces early engineering jobs, tech companies disperse salaries among a smaller number of people.
Gold star candidates are usually PhDs from a well-known computer science program focused on machine learning, such as Stanford, Berkeley, MIT, Waterloo, or Carnegie Mellon, and who, five years later, have landed on a top AI research team at a major company, said Thorne and Bai.
"Pretty much every client that Mark and I talk to would like to poach someone out of OpenAI or FAIR at Meta," Thorne said, referring to Meta's Fundamental AI Research group.
Fisher added that xAI is often flagged. Many top researchers also know one another, a major boon when assembling a team.
"More than specific skills, companies want people who live and breathe AI," Fisher said.
He said that this includes people who ship side projects, stay close to model updates, write code, speak at conferences, have a good social media presence, and write academic papers.
The AI wars fuel a newfound work intensity
The talent wars are a key symptom of "a level of urgency in tech that's been missing for years," Cordova said.
With every new AI model launched, a new wave of companies is born and dies, Bai said. The intensity is manifesting in more rigid return-to-office expectations and "founder mode" taking hold.
"They're bringing in some of this 996 culture," Bai said, referring to a widespread tech work culture in China of working 9 a.m. to 9 p.m. six days a week.
Bai added that some companies work seven days a week. "There are literally offices with bunk beds."
While a high-pressure atmosphere is synonymous with startup culture, Thorne said the pendulum has swung back in a major way after the COVID pandemic, and founders have become more forward about their demands.
"This is by far the most intensity that I've ever experienced from clients over the last decade," Bai said.
