Atomic just raised $260 million more to build and fund its own companies

Jack Abraham has a lot of confidence in what he’s building. Then again, you can’t be immodest or unsure of yourself if you’re going to bet exclusively on your own startups as an investor, which is precisely the model that Abraham’s San Francisco-based venture studio, Atomic, has followed since it was launched nine years ago.

It all started with $10 million of largely Abraham’s own money, capital he amassed by selling his first startup, a local shopping engine called Milo, to eBay in 2010 for $75 million. Abraham had dropped out of Wharton as an undergrad with $500,000 from a professor who believed that Abraham — whose father founded ComScore — would himself be a company-building machine.

The professor had good instincts. After selling Milo at age 24, Abraham spent more than three years building products inside of eBay and learning how to lead multiple teams before beginning to look outward, making angel bets, including on Uber and Pinterest, and, he says, spreading around some of his ideas. (Among these, he says, he “invented Postmates. I gave the founders literally the idea for the company; they were working on a B2B company at the time. I was fairly early on there; that helped spawn the whole food delivery thing.”)

He had so many ideas — hundreds, he says — that not long afterward, he created Atomic, later bringing on to the founding team Andrew Dudum, a Wharton peer who is also the son of entrepreneurs and who also dropped out of college to join the startup world. (Dudum’s first stop was a then-nascent startup backed by Sequoia Capital.)

At first, Atomic worked on one company. The following year, it worked on two. By 2018, the outfit had built out a team that could handle many of the back-end functions that startups need to thrive, from recruiting to accounting, and launched 10 companies. Impressed investors gave the firm $150 million to create even more startups.

By then, Abraham and Dudum had brought in two other general partners: Chester Ng and Andrew Salamon. Salamon didn’t stay long, eventually launching his own venture studio, Material, with Blue Apron founder Matt Salzberg in 2019. The same year, JD Ross, one of a handful of co-founders of the newly public company Opendoor, joined Atomic as a general partner.

The firm has picked up speed since. Indeed, at this point, Atomic has created “dozens” of startups — including roughly one per month last year, says Abraham. It also just closed on $260 million in new capital commitments, including from a prominent university that now serves as its anchor investor but would prefer not to be named publicly.

Citing “proprietary aspects” to the model, Abraham declines to explain how Atomic’s economics work, except to acknowledge that it operates in “more of a fund context instead of a holding company” where investors would essentially be buying stakes in Atomic itself.

Certainly, it’s easy to appreciate the enthusiasm of Atomic’s investors, including early backers like Peter Thiel and Marc Andreessen. Abraham and Dudum are both compelling storytellers, as we’ve witnessed firsthand in interviewing them at different times. The firm is also starting to see some exits.

One of Atomic’s creations, the telehealth company Hims, was taken public in January through a blank-check company in a deal that valued the company at $1.6 billion, and its shares have been rising since. As of this writing, the three-and-a-half-year-old outfit — run by Dudum, who is doing double-duty as Hims’s CEO and a general partner with Atomic — boasts a market cap of $2.9 billion.

Atomic also sold a voice-powered sales startup, TalkIQ, to the company Dialpad in 2018 for what Forbes reported at the time to be a “little under $50 million.” TalkIQ had raised $22 million altogether.

More exits are coming, suggests Abraham. “There are many companies we have that are now approaching the sort of growth and run rate where they have the ability to go public, even as soon as in the next year,” he says.

One of those eventual prospects is Replicant, an autonomous call center startup that has raised $35 million since its 2017 founding, including a $27 million Series A round led by Norwest Venture Partners back in September. Another Atomic startup, Homebound, a three-year-old home-building outfit that handles everything from financing to construction, has also enjoyed some momentum, as well as attracted $53 million from investors.

Though Atomic prides itself on “pressure testing” its ideas, not every startup has been a hit with users. A photo-sharing app called Ever was turned off soon after NBC reported that the photos people shared were used to train a facial recognition system — tech the company offered to sell to private companies, law enforcement and the military. (Ever was later refashioned into the enterprise company Paravision.) A sleep-tracking specialist, Rested, that Salamon was running was also shut down.

Meanwhile, ZenReach, a Wi-Fi marketing company that had collected at least $94 million from investors through 2018, laid off 20% of its employees that same year. A CEO who’d been brought aboard by Abraham and who was previously an operating partner with Atomic, has since moved on to a role outside of Atomic’s sphere.

If not all of its ideas set the world on fire, Atomic has no shortage of others.

Asked about some of the areas where he sees the most opportunity to innovate, Abraham quickly ticks off “healthcare, finance, education, real estate and other large industries where truthfully, when you’re inside them, you understand how broken they are, and they are broken up and down the entire stack.

“You study them,” he says, “and then you wonder: How is this possible this happened?”


Early Stage is the premier “how-to” event for startup entrepreneurs and investors. You’ll hear firsthand how some of the most successful founders and VCs build their businesses, raise money and manage their portfolios. We’ll cover every aspect of company building: Fundraising, recruiting, sales, legal, PR, marketing and brand building. Each session also has audience participation built-in — there’s ample time included in each for audience questions and discussion.