Counter-UAS start-up Yolo Machines: Romania’s bid to change air defence

Just one year ago, Romanian entrepreneur Lorand Minyo founded the start-up Yolo Machines. The company, based in Cluj in Transylvania, develops autonomous counter-UAS systems and is pursuing an ambitious goal: to fundamentally change air defence in Romania and Europe.

The driving force behind this can be felt directly on NATO’s eastern flank. Russian drones repeatedly penetrate the airspace of EU and NATO states – Romania is particularly affected. Only a few months ago, a Russian kamikaze drone that had veered off course after being hit over Ukraine struck a residential building in Galați. Two people were injured and the building caught fire. For Minyo, this is a clear warning sign.

At the same time, the entrepreneur sees the war in Ukraine as a driver of innovation. There, he says, new technologies and tactics are developed and tested at a pace that would hardly be imaginable in peacetime. “In this war, Ukraine has so far succeeded in developing innovations spontaneously and making optimal use of its military resources to counter the extremely costly Russian attacks,” says Minyo.

In the coming months, Yolo Machines intends to complete and test the first prototype of its autonomous counter-UAS system. In the following interview, Lorand Minyo talks about the lessons of the war in Ukraine, Romania’s role as a NATO frontline state, cooperation with the Ukrainian defence industry and why Transylvania of all places offers the best conditions for building a European defence tech start-up.

Mr Minyo, when and how did the story of Yolo Machines begin?
Yolo Machines was founded in 2025 in Cluj-Napoca as a programme of INC3 Ventures SRL. The starting point was a single observation about cost geometry: when a 400-euro quadcopter put a 20-million-euro main battle tank out of action for the first time, the economics of land warfare changed – and never returned to the old state. At the same time, Shahed-class loitering munitions began crossing borders in their hundreds. This also happened repeatedly in NATO airspace along Romania’s eastern border. We came to the conclusion that counter-drone defence is a layer of warfare that did not exist a decade ago, and that no European armed force is currently deploying it at the pace at which the threat is actually arriving. Yolo Machines was founded to build this layer. Secondly, we observed the asymmetry in human resources, which we address with our latest programmes.

Yolo Machines is one of 16 companies in the Techcelerator Vanguard programme. What does the programme involve, and how will it help the company?
What a programme at this stage does for us fits directly with our development path. We have fixed our architecture and are about to begin a seven-month prototype development that is tied to the closing of our seed funding. The most valuable things a techcelerator offers us at present are access to capital and programme partners, contact with players from the defence sector and the discipline of external milestones – all of this shortens the path from a fixed concept architecture to a demonstrated prototype. We are building a single system and nothing else, so anything that accelerates this one product line is worthwhile.

Your company specialises in autonomous anti-drone systems (anti-UAV). Could you go into more detail and tell us where Romania stands at European level in terms of production?
Our system consists of an integrated architecture with three elements. The M0THER is an optical carrier drone – a multirotor weighing less than 25 kilograms that passively detects and classifies threats with daylight and thermal imaging cameras without emitting signals that an adversary could target. As the second layer, the S0N is a single-use micro-interceptor: palm-sized, fitted with an air duct for impact resistance, fast enough to reach a target within seconds, and expended on impact – “you only launch once” is literally the name of the company and its first design requirement.

Are there any others?
The system is completed by the BAY, an autonomous ground station that recharges the carrier and reloads its interceptors in less than five minutes without ground personnel. A troop deployment consists of three M0THERs in overlapping patrols with twelve ready-to-launch S0Ns. No interceptor is launched until two of the three carriers independently agree – there is no command post and no cloud, nothing central that an adversary could attack. The same network can be scaled along a border in border mode. The fundamental design requirement behind all of this is cost: the interceptor must cost less than the drone it shoots down. As for Europe: countering unmanned aircraft systems (UAS) is regarded as a declared capability gap in more than 18 EU member states, and almost nobody is yet arming at the pace at which the threat is growing. We are still in the pre-prototype phase and are honest about that – but Romania has the technical base, and we are building our series production here in Cluj.

Most Romanian defence companies are based in or around Bucharest. Yolo Machines is based in Cluj-Napoca. Why there of all places?
Yolo’s location was chosen because of the pool of engineers. Our system relies on three disciplines – optics, embedded systems and machine learning – and in Cluj-Napoca we can hire specialists from all three fields locally. That is also why the series production line is planned in the same city and will not be moved to another location. Being outside Bucharest is not a disadvantage for us, because the specialists who actually develop this system are here. Cluj has always been a technology leader, and we are building on our own decades-long tradition in the city.

“Using an effector costing several million euros against a 30,000-euro drone may win the individual engagement, but it loses the war.”

How does the experience of the technological war in Ukraine influence the company’s vision, strategy and projects?
Directly. In Ukraine, the two threats that define our system are no longer merely theoretical. The first-person view (FPV) drone at troop level, which puts armoured vehicles out of action for a few hundred euros, and the loitering munition at long-range level, which covers thousands of kilometres to attack, for example, a substation. Both share one characteristic: they are cheap enough to be deployed in quantities that no defender can match on a per-unit basis. That is the entire premise of our doctrine, and that is why we refuse to apply traditional air defence concepts to this problem. Using an effector costing several million euros against a 30,000-euro drone may win the individual engagement, but it loses the war. Our threat database is trained on the signatures of modern FPV drones and loitering munitions, and our entire cost geometry is geared to the exchange that the war has made inevitable.

Are you considering partnerships – now or in the future – with similar companies from Ukraine, including those with combat experience?
We are in the pre-prototype phase and are open to capital and programme partners as we move into the laboratory build phases and the validation phase. Combat-proven operational feedback is of course valuable to anyone developing in this category, and Ukraine has the deepest experience in this field. We are therefore in direct contact with the leading companies that have first-hand operational experience.

How do you assess the pace at which Romania is adapting to the new military-technological realities? Is the country prepared for Russian threats, and how well equipped is it in the field of anti-drone systems in particular?
I would look at this at European level, as that is where the public information is clearest. Countering UAS is named as a priority area that more than 18 EU member states want to cover jointly, and in NATO airspace there have been repeated incidents involving Shahed debris along Romania’s eastern border since 2022. So the threat here is not abstract; it has literally landed. Frankly, no European armed force, Romania included, is yet able to provide counter-drone defence at the pace at which the threat is actually arriving. It is precisely this gap that is the reason companies like ours exist. What I would say specifically about Romania is that the country has the technical base to be part of the solution – and that is exactly what we are betting on by producing here.

“no European armed force is yet able to provide counter-drone defence at the pace at which the threat is actually arriving.”

What are the plans for the coming years with regard to the production of anti-drone systems? Will you expand production capacity?
Our roadmap is public and divided into phases. The architecture is fixed; the next step is a seven-month test build of carrier platforms, BAYs and interceptors, which depends on the closing of our seed funding. This includes laboratory validation in 2027, field validation in 2028, a live intercept demonstration in 2028 or 2029 and series production from 2029. Each phase is a prerequisite for the next – we do not skip any steps. When we ramp up manufacturing, the production line is planned for Cluj-Napoca, with the supply chain deliberately located within the European Union. So yes, we intend to expand production. But only after completion of the validation phase, which we are deliberately not rushing.