Evos raises $9M to deploy autonomous operators in physical industries

The New York startup targets freight, manufacturing and logistics SMEs, where retiring experts take irreplaceable institutional knowledge with them.

By Disrupts Editorial Team Tue, 27 October 2026 AI level 2 of 5

AI level 2 of 5: drafted by our AI editorial assistant from the company's announcement; fact-checked, edited and signed off by Mark Walker, Editorial Director. What the levels mean

A brightly lit, sterile cleanroom features two long, parallel lines of automated manufacturing machines with screens and indicator lights on a reflective white floor.

Evos, a New York-based agentic-operations startup, has emerged from stealth with a $9m pre-seed round co-led by Uncork Capital and Field Ventures, with Crosslink Capital, Utopia Capital Management and Sandwith Ventures also participating. The company is targeting a corner of the economy that enterprise software has historically underserved: small and mid-sized businesses in freight, manufacturing and other physical industries, where the most valuable operational knowledge often exists only inside the heads of long-tenured workers.

The pitch is pointed. According to Evos, 99.3% of US motor carriers operate fewer than 100 trucks, and 93% of US manufacturers employ fewer than 100 people. These businesses run on tacit expertise accumulated over decades, and that expertise is retiring faster than it can be replaced. Evos's platform, branded Workhorse and due for formal launch in the coming weeks, is designed to capture that institutional knowledge and encode it into autonomous operators that handle discrete operational functions end to end.

Learning from behaviour, not prompts

What distinguishes Evos from conventional AI tooling is its training model. Rather than requiring workers to write down processes or construct prompts, the platform learns from observed decision patterns: what a team's most experienced people actually do, case by case. An operator only moves to autonomous execution once its decisions consistently match the team's own, task by task and scenario by scenario. Novel situations revert to human review, and each human decision feeds back into the model.

The company reports that operators reach roughly 82% decision accuracy across their deployments, handling around 70% of work without human intervention. A freight-brokerage client, MyFreightWorld, has used an Evos operator to audit carrier invoices against its transport-management system; in two months the operator made nearly 3,500 audit decisions at 97% accuracy and closed out more than $600,000 of invoice value in five weeks.

"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. "An Evos operator learns the job from these people, improves on it, and proves it can match them before it takes on any work by itself."

The company passed $1m in annualised recurring revenue four weeks after going live, which is the substantive proof point backing its valuation at this stage.

The convergence angle: agentic AI meets the physical-economy labour gap

The Evos raise sits at the intersection of two macro shifts that investors across sectors are pricing simultaneously. First, the structural labour shortage in physical industries is accelerating adoption of AI operations tools in markets that were previously too fragmented and process-variable to automate economically. Second, the maturation of agentic AI, systems that own a function rather than respond to a prompt, is shifting the unit of automation from the individual task to the operational role.

For capital allocators, the implications extend beyond logistics software. If platforms like Evos can genuinely encode and replicate institutional expertise at scale, the addressable market reaches across any physical-economy vertical where knowledge transfer is a constraint: food production, utilities, field services, construction. The approval-gate model Evos uses, where humans validate decisions that then become training data, also offers a practical answer to the enterprise-AI adoption question: how do you get a risk-averse SME to trust an autonomous system on day one? You do not. You earn autonomy incrementally.

The deeper strategic question is who owns the operating layer for physical industries. Large ERP vendors such as SAP and Oracle have sold record-keeping systems to these businesses for decades. Evos is positioning itself one layer above: not tracking work, but doing it. That framing, if it proves out at scale, threatens the adjacent professional-services revenue that system integrators and consultants have historically earned from implementation and workflow design. The $9m pre-seed is a modest opening bet on a very large reordering of how physical businesses get run.