AI startup Evos has raised $9 million in pre-seed funding to develop autonomous AI operators for small and midsized industrial businesses, including freight brokerages. The company announced the funding on Wednesday as it seeks to automate routine operational tasks without requiring customers to replace their existing software systems.
The financing round was co-led by Uncork Capital and Field Ventures, with additional participation from Crosslink Capital, Utopia Capital Management, Sandwith Ventures and angel investors.
Evos also reported reaching more than $1 million in annual recurring revenue just four weeks after launching. However, the company has not disclosed its valuation or the number of customers currently using its platform.
AI Designed to Learn How Businesses Actually Operate
Rather than functioning as conventional AI assistants that depend on employees to issue instructions for every task, Evos is developing software operators capable of learning established workflows and eventually carrying out assigned business functions independently.
The platform observes how employees make decisions within a company’s existing systems. Once its decisions consistently align with those of experienced workers, an operator is designed to take over the relevant tasks without requiring continuous human prompting.
The approach is aimed particularly at smaller industrial companies, where essential operational knowledge often resides with long-serving employees rather than in formal documentation or standardized procedures.
“The most capable system in many of these businesses is the person who has been there 20 years, and what they know has never been written down,” said Urav Shah, founder and CEO of Evos.
The company argues that this dependence on individual experience creates an opportunity for automation, particularly among businesses that lack the technical resources to undertake complex digital transformation projects.
According to figures cited in Evos’ announcement, 99.3% of U.S. motor carriers operate fewer than 100 trucks, while 93% of U.S. manufacturers employ fewer than 100 people. For these businesses, the cost, implementation time and technical demands associated with traditional automation can represent significant barriers.
Evos says its platform is designed to avoid those obstacles through a self-service deployment model.
“The platform is fully self-serve: work that would take a consultant or an implementation partner 9 to 18 months to deliver, an Evos operator takes on in 24 hours from discovery to deployment,” the company said.
MyFreightWorld Employs AI for LTL Invoice Audits
One example of how the technology is being put to work in day-to-day logistics operations is freight brokerage MyFreightWorld.
The company uses an Evos operator to check LTL carrier invoices against their transportation management system, looking for discrepancies in the billing information.
Evos says the operator made almost 3,500 audit decisions over a two-month period with a stated accuracy rate of 97%. In the startup’s most recent five-week period of data, the system detected billing discrepancies of more than $600,000.
There is an important distinction to be made about that figure: the $600,000 is the value of discrepancies found between MyFreightWorld and its end customers’ invoices. This is not money recovered, refunds confirmed or savings achieved.
Evos also hasn’t explained how it calculated its claimed 97% accuracy rate, or identified the sample against which that performance was measured. So the figures are still company-reported results, not independently audited performance indicators.
However, the application demonstrates the potential use of AI operators within freight brokerage workflows, where invoice validation may require checking multiple charges and cross-referencing carrier bills with transportation records.
LTL Billing Errors Build an Opportunity for Automation
It is especially critical to audit invoices in the less-than-truckload segment where freight rates can vary based on additional services, fuel surcharges and classifications of shipments.
Errors in accessorial fees, fuel surcharges and freight classification can cause a carrier to invoice different amounts than a customer expects to pay.
An idea of the scale of these problems can be seen from an analysis of invoices reviewed by AFS Logistics in 2024. The company found 11,881 errors in nearly 250,000 invoices reviewed.
42.8% of errors found were incorrect accessorial charges. Fuel surcharges were 14.4 percent and freight misclassification was 12.3 percent.
Although these results are specific to AFS Logistics’ review, these should not be interpreted as an error rate across the entire LTL market. But they are indicative of the range of billing problems that invoice auditing processes might have to deal with.
Automating these checks can help spot inconsistencies that would otherwise require a lot of manual effort for high invoice volume freight brokers. But the economic impact depends on whether the anomalies detected can be verified and then fixed.
Logistics Companies Vie to Cash In On AI
Evos is entering a market where logistics companies are still working out how to use AI and waiting for real proof of its monetary value.
A report published by Boston Consulting Group in August 2026 found that only 13% of executives at 30 leading global logistics companies surveyed said AI was delivering measurable financial results.
Still, two-thirds of respondents said they have a dedicated budget for AI and a strategy for the technology.
The results suggest there is still a gap between investment in AI and the ability to measure business outcomes from that investment.
That challenge can be that much starker for smaller firms given the cost of implementation and operational complexity. Systems needing long integration projects, technical champion support or major workflow changes are tough to justify, especially for resource-constrained businesses.
Evos is wagering that it can democratize automation by teaching A.I. operators to operate within the constraints of current software systems and processes. The model is built to avoid long deployments and allow companies to keep using the tools they already have.
The ability of the operators to be reliable, the complexity of the workflows they manage and the ability to show measurable financial benefits will determine how well this approach can deliver consistent results across different industrial businesses.
Evos’ growth will depend on showing its worth
Evos’ platform development will be funded in part by the $9 million pre-seed round as demand for real-world uses of AI for industrial operations grows.
This is the defined use case of what it does with MyFreightWorld. Reviewing LTL invoices, comparing charges with transportation records and looking for any possible billing discrepancies.
But the distinction between identifying an error and offering financial restitution is central to assessing the impact of the technology. The startup has not disclosed how much money its system has helped recover and its claimed accuracy rate has not been independently verified.
For logistics companies considering similar tools, the critical question isn’t whether artificial intelligence can automate rote tasks, but whether it can do so reliably and deliver tangible operational or financial benefits.
“We are seeing growing uptake of AI in freight and industrial operations, and solutions that can deliver provable results with little of the implementation burden may have an edge, particularly at the smaller end of the spectrum where the cost and complexity of traditional automation projects can be prohibitive.
