Hiring’s a lot like dating. 

It starts with two people sitting down and having a conversation. Just maybe, it leads to another conversation. There has to be mutual interest for it to work, but there’s often a pursuer and a pursued. Eventually, if all goes well, both parties make a decision that affects their daily lives. Obviously, there are differences (last I checked, most people have more jobs than they have life partners), but in essence—these are highly competitive and deeply interpersonal markets. And in startups right now, particularly for go-to-market and engineering talent, competition has gotten wild. 

“It’s a hiring free-for-all,” said Rhys Hughes, GV executive talent partner. “There’s, of course, the [AI] bonus hilariousness out there, the great talent auction, and all the things making it competitive. Right now, with our portfolio companies, I’m spending a ton of time with them on this proactive approach. If a founder says to me, ‘I’m open to a rec,’ I say: ‘You’re already late. You should already know three to four people.’”

I spoke to six founders and VC firm talent leaders, and there was agreement—we’re in a new era of hiring for startups. Once, it was a “growth at all costs” game, characterized by soaring headcounts. Those days are over, replaced by a push for companies to run leaner teams of exceptional people—who are often already established in their careers. 

“The biggest thing I’ve noticed is that people care a lot more about talent density than about headcount growth now,” said Atli Thorkelsson, VP of talent network at Redpoint. “I think what’s feeding into it is this: There’s more you can automate now, especially within engineering and all kinds of other functions. You can automate a lot of junior-level work, so what you’re really looking for is higher-level thinking. There’s a culture, or ethos, right now—whether it’s true or not—that building the next really cool AI product may be harder than building the great SaaS products of the past.”

Thorkelsson and Redpoint released a report this year that showed that while demand for mid-level and senior talent is high, demand for entry-level talent is declining rapidly across startups. This is naturally linked to AI—as the technology develops and changes quickly, track records and adaptability matter more than traditional credentials. 

“For a while, I was really focused on: Have you done it before?” said Garrett Lord, cofounder and CEO of Handshake. “Now, after hiring hundreds and hundreds of people, I’ve really learned it’s more about learning agility, judgement, and work ethic than about you spending four years at Meta.” 

Lord has chased potential hires across the globe—even recently meeting a prospective executive in Dubai. He says the competition for AI talent is as intense as conventional wisdom right now suggests. 

“If you’re remotely connected to or have worked at one of the AI companies, there are seed funds that will just write you a check on the spot,” said Lord. “You could own your own company. You could go to a big company, and cash in on public [restricted stock units]. Not Meta-style, but significantly more money than many folks have ever seen.”

So, if you find a great candidate, don’t hold back.

“My advice to founders is that when you find a candidate you like and that first meeting goes well, lean into them,” said Paula Judge, VP of talent at Accel. “Start texting them immediately, and build the relationship. Let them know how you’re going to work with them. Act like you’re working together. Show them what it’s going to be. That’s really the dance. That’s how people and companies differentiate. People do say ‘I took this job for less money because I really liked the people.’ Candidates are looking to do their best work.” 

To compete, you have to start early, said Eléonore Crespo, cofounder and co-CEO of Pigment. 

“I think of it like what sports teams do to find the future stars,” she told Fortune via email. “They look at 16-year-olds that show promise, and they keep tabs until the moment is right. I try to do the same every day.”

It’s an essential point, said GV’s Hughes—as in sports recruiting, the long game matters. 

“The best founders are building relationships with talent long before they need them,” he said. “So when they do give them that call, that person will take that phone call, because you’ve invested the time. That’s going to be the currency that matters most for our companies as they hire and scale.”

And, of course, as in dating, hiring is only the beginning: Once someone’s in, it’s tougher and more important than ever that they stay, said Jen Holmstrom, Notable Capital partner. 

“Once you get them in the door, how do you ensure that particular talent is going to stay invested?” said Holmstrom. “Do they feel they have autonomy, that they have room to run, that they’re trusted? You have to create an environment where you retain them, keeping them with you on the journey. And that responsibility sits with the founder. It sits with the leadership team. And it’s not easy to do. It’s not to be underestimated.” 

See you Monday,

Allie Garfinkle
X: @agarfinks
Email: [email protected]
Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

© Getty Images

The soaring valuations of the early 2020s are, finally, coming back to earth.

This week, PitchBook data revealed that 15.9% of venture-backed deals in 2025 so far have been down rounds, marking a decade high. Additionally, almost every major IPO listing in Q2 hit the public markets below its peak valuation, the data from PitchBook adds. Some examples include MNTN (at IPO, valuation was down from $2 billion to $1.1 billion), Circle (dropping from $7.7 billion to $5.8 billion), Hinge (valuation at IPO was $6.2 billion, down from the $23 billion high), and Chime (going public at $9.1 billion from a $25 billion peak valuation). 

AI continues to be a bright spot in many ways—but isn’t entirely exempt either, as 29.3% of down rounds were in PitchBook’s broad AI and machine learning vertical. Of course, the biggest names in AI—like OpenAI, reportedly heading towards a $500 billion valuation, and Anthropic, reportedly raising at a $170 billion valuation—continue to hit eye-popping levels. And lower on the food chain, AI is still consistently valued at a premium, with PitchBook reporting that median Series B step-up for AI startups is 2.1x, well above the median of 1.4x that all other categories fetch. 

The IPO market is, some would say, back. (I think it pretty much is, but also has been for a while, for those with the stomach for it.) In Q2, venture-backed startups in the U.S. generated $67 billion in exit value, PitchBook said—the highest since the last quarter of 2021. But here’s a sobering fact: There are still lots of unicorns out there, so distributions back to VC firms and by extension their LPs are limited. 

An unsurprising but ice-cold (and very down-to-earth) number to leave you with: The unicorns that have made their public debut this year comprise a mere 1% of all U.S. unicorns.

See you tomorrow,

Allie Garfinkle
X: @agarfinks
Email: [email protected]
Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

© Bloomberg/Getty Images

Rebecca Lynn doesn’t believe in first-mover advantage.

“Never be a first mover,” says Lynn. First movers bear the burden of proving that a market exists, she says. “You’re paying to get consumer awareness of that market. You’re fighting this uphill battle. You’re spending a ton of money to prove out the market—and by the time you finally prove it exists, you have an aging tech stack. Why not let somebody else break that pick? You watch where the puck is going, and then come in with a better product and better marketing as a second mover.”

It’s advice Lynn has gleaned from decades of experience on the operations and investing sides of the table. Lynn never planned to be a VC—a first-generation college kid from the Midwest, she spent her early career in the late ’90s at Procter & Gamble. In 2007, right before the Great Recession fully materialized, she ended up taking her first venture gig at Morgenthaler Ventures. The very first term sheet she gave out was for financial services company Lending Club in 2009. 

“Always keep an eye on the macro,” Lynn says. “So, when Lehman crashed, that’s what made Lending Club a great deal. Without the crash of Lehman, we wouldn’t have invested in Lending Club. But there was this macro opportunity where suddenly their product just nailed it. They were doing peer-to-peer lending and, before the crash, it was a good idea. Then, the market goes down in flames, and even if you have an 800 credit score, you can’t get a loan. It created this opportunity where they had a clear runway ahead of them.”

In 2014, Lending Club would become the biggest tech IPO of the year. Since then, she’s also been an early investor in Doximity, Luminar, FutureAdvisor, Check, and Casetext. Now, twelve years after cofounding Canvas Ventures (spun out from Morgenthaler Ventures in 2013), Lynn is spinning out on her own, launching new firm Canvas Prime as a solo GP, she exclusively told Fortune. 

“We got a lot of guidance from our LPs that they liked my strategy and results,” she says. “And I had really, from a top-down level, wanted to focus more for years. Because focus—that’s the Kool-Aid we tell our entrepreneurs to drink: Focus, focus, focus, execution, execution. Make a decision, and make it fast. Because not making a decision is also making a decision.”

Lynn is keeping the Canvas name, as she’s retaining board seats and continuing to support her current portfolio companies, including Savvy Wealth, Airvet, and Marvin. In some ways, Lynn is part of a broader trend in venture, of solo GPs who have a deep well of experience, leaving older or larger firms behind. Her strategy for Canvas Prime is defined: The firm concentrates on a few sectors—primarily fintech, digital health, and AI—where she has proven expertise and a track record, plus a concentrated, high ownership portfolio. They invest in 12-15 companies a year while taking significant ownership stakes of 15% to 25% to maximize returns. 

“I think the days of the mid-market, generalist fund are absolutely gone,” says Lynn. “They don’t work. Smaller funds outperform, and here’s a number: Specialized funds have 37% higher returns than generalist funds… And when we did the bottom-up analysis on where we made money for our investors, it was blindingly clear we made money in fintech, digital health, and AI—and that was it.”

In a lot of ways, Canvas Prime is about remembering that it’s important, and necessary, to say no. 

“You just can’t be an expert at everything,” Lynn told Fortune. “What we tell the CEOs all the time is that knowing what to say no to is the most important skill set you can develop. So, it’s all about having focus and clarity on what we need to cover, how we need to cover it, who we need to hire, and what we need to build out to be helpful to our companies.”

Term Sheet Podcast, Episode 3… This week, we’re trying something new on the Term Sheet Podcast—pairing an episode with an essay! In my Term Sheet Podcast interview with Rebecca Lynn, we discussed her roots in the Midwest, her path to Silicon Valley, and what she’s learned as an entrepreneur turned venture capitalist. Plus, my take on the Nvidia-AMD-Trump deal and newly minted AI billionaires. Listen here.

See you tomorrow,

Allie Garfinkle
X: @agarfinks
Email: [email protected]
Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

© Canvas Prime

Manufacturing dates back about 2.5 million years, when early hominins fashioned stone tools. Since then, it’s been an industry formed and reformed by technology. 

And as AI has become the axis on which all technology turns, it makes sense AI would make its way into factories of all kinds—but startups focused on AI in industrial applications are still rarer than you may think, said Devin Bhushan, founder and CEO of Squint. 

“For whatever reason, it can still feel like a forgotten industry by a lot of the tech world,” said Bhushan. “Even today, it can be hard to, say, get together for an industrial founders dinner. There’s not a group I can meet up with. There’s not that many people [in Silicon Valley] serving it. When we look at our competitors, our biggest competition is binders—physical binders.”

Bhushan’s Squint uses AI and augmented reality (AR) technology to capture and document the expert knowledge of experienced factory operators. The idea is this: to use AI to turn everyday workers into experts and to make the invisible backbone of the economy smarter, safer, and more efficient. 

“We’ve basically built a manufacturing intelligence platform… that uses AI to extract that knowledge from operators who have been in the environment for 30 years. They have all this stuff up here,” said Bhushan, gesturing to his head. “We extract it by watching them do the work. Then AI auto-documents it and creates procedures. From there, we can enable anyone to be an expert on pretty much any task.” 

Squint has raised a $40 million Series B at a valuation of $265 million, Fortune has exclusively learned. The Westly Group and TCV led the round, with participation from existing investors Sequoia Capital and Menlo Ventures. Bhushan founded the company in 2021, and the company’s current customers include PepsiCo, Michelin, and Ford, tens of thousands of operators at hundreds of factories. The round comes at a time when “every industrial company in the world is searching for ways to keep their workforce… ahead of the curve,” said Steve Westly, founder and managing partner at The Westly Group, via email. This is, of course, both because of Trump-fueled geopolitical tensions rising around manufacturing and the rise of AI. 

“In our factories, there’s more desire to insource,” said Sequoia partner Jess Lee. “There’s more pressure and incentive to move manufacturing back here. For geopolitical reasons too, there are folks in some of the essential critical industries to manufacture here… In order to make that possible, factory productivity really matters.” 

To Lee—who met Bhushan when both were at Yahoo, he as an engineer and she in the aftermath of the company’s acquisition of her startup Polyvore—the realm of physical work is vast and filled with untapped potential. 

“The world of tech is so white collar-focused, so much about digital work,” Lee told Fortune. “But the world of physical work is enormous, and it supplies us with all the things that power even the digital industry. I think tech has barely scratched the surface of modernizing that, and I think that’s where some of the next biggest tech companies are going to be created.”

Bhushan once built features for Yahoo’s fantasy sports platform, where he also met Tim Tully, partner at Menlo Ventures, who led Squint’s pre-seed on the idea that “there was a thirst for richer experiences in manufacturing,” a thesis that he says via email has “rung true.”

Though he’s moved on from building the platform millions play fantasy sports on, Bhushan keeps sports metaphors close. 

“We have this saying inside the company—it’s our number one value: ‘Dunk the three,’” he said. “Do the thing no one has ever done. No one’s ever dunked a three in an NBA game, right? But when someone dunks a three, it’ll become a stat. Then, more people will dunk threes. That’s how it works.”

But no one can dunk a three, I point out.

“Well, Michael Jordan did in Space Jam,” Bhushan laughed. “He extended his hand, and dunked the three. That’s what we’re trying to do with our tech.”

See you tomorrow,

Allie Garfinkle
X: @agarfinks
Email: [email protected]
Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

© Squint

A big week for OpenAI is a big week for AI. 

And this week saw the Sam Altman-led company unveil its long-awaited GPT-5, along with a sweeping U.S. government partnership. There’s also a key number now in the mix: $500 billion, the reported new valuation OpenAI is set to hit amid a share sale. 

That half-a-trillion number—up from $300 billion—is stunning on many levels. I’ll spare you the obligatory market cap comparisons (“that’s the equivalent of 10 Ford Motor companies!”). Wild though the prospective valuation may seem, it also poses important questions about where we are in the cycle of this AI boom. Venture funding, overall, isn’t booming as much as you’d expect. According to new Crunchbase data, global venture funding hit $29.7 billion in July, down from $43 billion in June. AI, of course, continues to attract most of the action, accounting for 37%, or $11 billion, of the July funding numbers.

And for the top-tier AI deals, valuations are breathlessly high, and that’s not just the case for OpenAI: Anthropic’s reportedly heading towards $170 billion, Perplexity’s reportedly at $18 billion, and Elon Musk’s xAI (yes, also reportedly) is chasing a valuation of $200 billion. This is all happening quickly, said Nnamdi Okike, cofounder and managing partner of 645 Ventures. 

“Many companies are raising huge amounts of money in very short periods of time,” he told Fortune. “Even when there’s fundamental substance behind the rounds, that means there’s typically some dislocation. If you zoom out, how many winners are there really going to be? That’s the really important question. With a rising tide, what typically gets missed is that—in most markets—there are only a couple winners. It might be the case that these fast-rising valuations are justified for one or two of those winners. But it’s certainly not the case that rapidly increasing valuations are justified across the board.” 

Okike does think OpenAI could be a long-term winner. But he’s also a history buff, noting that broadly we’re seeing classic signals of market euphoria and potential bubbles.

“History teaches us that when valuations go up very rapidly, when round sizes and valuations are higher than they’ve ever been, those signs indicate euphoria,” said Okike, who sees echoes of the Dotcom bust and the search engine wars. “They indicate potential bubble-like qualities, and that investors should be very careful when rounds and prices increase as quickly as they’re increasing.” 

See you Monday,

Allie Garfinkle
X: @agarfinks
Email: [email protected]
Submit a deal for the Term Sheet newsletter here.

Joey Abrams curated the deals section of today’s newsletter. Subscribe here.

This story was originally featured on Fortune.com

© Chip Somodevilla—Getty Images