Everful raises $48.5M to let AI agents pick your factories
Vertex Ventures Southeast Asia & India and January Capital are backing a San Francisco startup betting that finding and vetting factories is a software job, not a relationship one.
Everful, a San Francisco startup whose AI agents find and vet factories and manage orders for retailers and brands, has raised a $48.5M Series A led by Vertex Ventures Southeast Asia & India and January Capital, and says it is on track for $100M in annualised revenue by year-end. For operators and investors, the round tests whether sourcing, long a relationship business, can be run as software, and whether a run-rate claim of that size survives contact with a closing year.
Everful has raised $48.5 million in a Series A, and the San Francisco startup is attaching a number to the pitch that will get attention in any boardroom: it says it is on track for $100 million in annualised revenue by year-end. The round was led by Vertex Ventures Southeast Asia & India and January Capital. The product is a set of AI agents that find and vet factories and then manage orders for retailers and brands. That is the entire announcement, and it is worth unpacking, because both the raise and the revenue claim say something about where the unglamorous, enormous business of physical product sourcing is heading.
Sourcing is the least glamorous part of building a product and often the most expensive to get wrong. A brand decides it wants to sell a jacket, a lamp, a phone case. Someone then has to find a factory that can hit the target price, the quality bar, the lead time and whatever compliance standard the buyer's market demands. Historically that work runs through human sourcing agents, trade shows, referrals and long email threads, with quotes tracked in spreadsheets and trust built over years. It is relationship-heavy, opaque and slow, which is exactly the profile of work software companies have spent the last decade trying to eat. Everful's bet is that AI agents can do the finding and the vetting, and then stay on to manage the order once the factory is chosen.
The lead investors are a tell about where the supply base sits. Vertex Ventures Southeast Asia & India invests across a region that has become a primary destination for manufacturing capacity serving Western brands, and January Capital backs technology companies across Asia-Pacific. For a company whose core asset is knowing which factories can actually deliver, having investors with feet on the ground in those markets is not a vanity line in a press release. It is distribution and diligence in one. It also hints at the geography of the opportunity: as brands spread their manufacturing across more countries, the number of factories a buyer must evaluate goes up, and the value of a system that can screen them goes up with it.
Now the number that will get quoted most. Annualised revenue is a run-rate, not an audited annual figure. It typically takes a recent period of revenue and multiplies it out to a full year, which means it is a snapshot of momentum rather than a record of what has actually been collected over twelve months. Companies use it because it is the fastest way to show trajectory, and investors read it knowing exactly that. The relevant comparison here is scale: a $48.5 million raise against a claimed $100 million run-rate by year-end implies a company being priced for growth rather than for current profitability, and a company that will be expected to prove the run-rate holds once the year actually closes.
The hard part of this business is not matching. It is the failure modes. Orders arrive late, or wrong, or below the quality bar, and someone has to own the problem, negotiate the fix and decide who eats the cost. Human sourcing agents earn part of their fee on exactly that. An AI agent that can find and vet a factory and manage an order is useful; an AI agent that can absorb a dispute at 2am without the buyer losing a season is a business. That is the question every investor in this category is underwriting, and it is the question Everful will have to answer as its order volume grows.
The second-order effects run in a few directions. For retailers and brands, cheaper and faster sourcing widens the set of companies that can credibly manufacture a product, which matters most to smaller brands that could never justify a full-time sourcing team. For the human agents who have brokered these relationships for decades, the work shifts from finding to adjudicating, a smaller and more senior job pool. For factories, being legible to software becomes a marketing requirement rather than an option. And for the platforms that already sit between buyers and suppliers, a startup claiming a $100 million run-rate is a direct shot at a layer they assumed was theirs.
What to watch next is unglamorous and specific. Whether the run-rate survives an audit of time. Whether the company publishes customer names or stays quiet about them. Whether the agents handle compliance and audit paperwork, which is where sourcing liability actually lives, or stop at the match. And whether the next round is raised at a valuation that treats the $100 million as a floor rather than a peak. For operators in the category, the lesson is simpler: the parts of the supply chain that felt too relationship-driven to automate are being automated anyway, and the money is already moving.
This story's Key Insights and Take-aways are locked.
Create a free account to unlock Executive Actions for one credit.
Register to UnlockAlways free for Executives Club members. Join the Club
More in Technology
Anthropic mathematician claims Claude helped crack the Riemann hypothesis
If the proof holds, AI has toppled math's most famous open problem; if it does not, it is a case study in announcing breakthroughs before anyone verifies them.
AI's $4 Trillion Wipeout Erases Nasdaq's Gains for the Year
A brutal week took roughly $4 trillion off the AI sector and left the Nasdaq flat for the year, and the damage ran the full length of the supply chain.
Anthropic mathematician says AI proved the Riemann hypothesis. Verification will take time.
A single unverified claim from inside Anthropic now sits against more than 160 years of open mathematics, and the gap between asserting a proof and surviving peer review is the entire story.


