Justin Ernest has put almost $500 million into some of the hardest private companies to access without building a traditional venture fund. That says a lot about where power in startup investing is moving.
The interesting thing about Sabertooth VC is not only that Justin Ernest found his way into names like Anthropic, Anduril and SpaceX. It is that he did it without the usual flagship fund, the long limited partner roadshow, or the standard venture structure that has defined Silicon Valley for decades.
Instead, Ernest has used special purpose vehicles, or SPVs, to assemble capital deal by deal from a network of technology executives and family offices. That may sound like plumbing, but in venture capital the plumbing matters. It decides who gets access, who takes fees, who appears on the cap table and who gets to participate before a company is too large or too expensive for ordinary investors to touch.
For founders, this is also more than a financing detail. A stack of small investors can create noise, paperwork and future governance friction. A properly organized SPV can turn that crowd into one line on the cap table. That makes the structure appealing to companies that want capital without turning every round into a shareholder management problem.
Traditional venture capital is built around patience. A firm raises a fund, commits to a strategy, collects management fees and then spends years deploying the money across a portfolio. That model still works, especially for early-stage investing where judgment, support and follow-on reserves matter. But the hottest private companies now stay private longer, raise enormous late-stage rounds and give very few outsiders a clean path in.
That is where Ernest's model fits. An SPV does not need to promise exposure to a whole portfolio. It can be built around one company, one allocation and one narrow investor appetite. If the network is strong enough, capital can come together quickly. That changes the fundraising rhythm from a long institutional campaign into a sequence of focused transactions.
This is why the model can feel uncomfortable for established venture firms. Their economics depend on committed funds, ownership targets and the idea that access flows through the firm. SPV-heavy investing pulls on a different thread. It says access can come from relationships, secondary markets, founder trust and a pool of wealthy backers who would rather choose a specific company than commit blindly to a decade-long fund.
There is a practical reason this is happening now. Companies such as Anthropic, Anduril and SpaceX are not just private startups in the old sense. They are institution-scale businesses with strategic importance, huge capital needs and investor demand that often exceeds available room. As Business Insider recently reported, Anduril raised $5 billion in a round that valued the defense technology company at $61 billion, a figure that shows how large these private rounds have become before public investors get a look.
Founders are choosing structure as much as money
The founder side of this story is easy to miss. Capital has become abundant at the top end of the market, but useful capital is still scarce. A founder deciding whether to accept an SPV is really asking whether the organizer can bring relevant people, clean execution and low distraction.
That is why a network of tech executives can matter. An investor base made up of operators, founders and family offices may not look like a traditional venture firm, but it can still carry strategic value. If the company is Anthropic, the useful person may understand enterprise AI adoption. If it is Anduril, the useful person may know defense procurement or hardware scaling. If it is SpaceX, the useful person may simply be patient enough to hold through long cycles that public markets often struggle to price.
The risk is that SPVs can become access wrappers with little responsibility after the wire clears. A good venture firm usually wants to protect its reputation over multiple rounds and multiple companies. A loose SPV ecosystem can be more transactional. Founders will have to separate investors who can help from those who only have a distribution list and a fast subscription process.
For investors, the trade-off is just as clear. A direct SPV into a famous private company can feel cleaner than paying fees on a blind-pool fund, but it concentrates risk. There is no portfolio construction doing quiet work in the background. If the valuation is stretched, if the liquidity window slips, or if the company changes direction, the investor owns that specific exposure.
That does not make the model weak. It makes it honest. SPVs are not replacing venture capital in every part of the market. They are taking share in the places where the old venture model is least efficient: late-stage access, oversubscribed rounds and companies whose names carry their own demand.
The larger message is that private markets are becoming more modular. Founders can choose between a fund, a strategic investor, a family office, a secondary buyer, an SPV organizer, or some mix of all of them. Investors can choose between broad exposure and specific bets. The winners will be the people who understand that access alone is not enough. In this next phase, structure is part of the product.
Also read: ASML's record run still leaves investors arguing it looks cheap • Orbital raises $5 million to test AI data centers in space • Palantir is taking Sadiq Khan's Met Police veto to court
