Field Notes is a recurring series from DK NEJET. Each entry takes one real problem from a contested electronic-warfare environment and follows it through the loop – from the operator who hit it, to the engineer who answered it, to the version that flew next. This is the first.
A few weeks ago, one of our systems stopped doing what it was supposed to do.
The specifics are not the point, and most of them stay where they belong — with the operators and the engineers who hold them under NDA. What matters is the shape of it. In the field, the adversary had changed something. A technique that had worked the week before no longer worked the way it should. An operator noticed, in the middle of a job, that the behavior was wrong.
In a conventional defense supply chain, that is where the story ends for months. The observation gets written down, routed, reviewed, and scheduled. Somewhere downstream, eventually, a fix is certified and sent to the field. However, by the time it reaches its destination, the opposing party has already moved twice more.
That is not how it went. The operator’s report reached an engineer the same day. The support engineer understood the problem because the engineer had stood where the operator was standing – that’s one of the advantages of employing veterans (not exclusively, but they do bring their share of experience and expertise). A change was built, tested against the actual conditions – not a clean lab approximation of them – and pushed back out. The gap between “this is broken” and “this is fixed and flying” was measured in days.
That gap is the entire company.
What we actually build
It is tempting to describe DK NEJET by its hardware, because hardware is visible. We make FPV platforms, interceptors, and strike drones, several of which are NATO-codified, approved, and fielded by three national ministries of defense. They are good. They work.
But the platforms are not what makes us different. The loop is.
not months
An operator hits a problem under electronic warfare, mid‑mission. A technique that worked last week no longer works the same way.
We build the software layer that lets autonomous systems keep functioning when communications are degraded, GPS is unreliable, and an adversary is actively working to sever the link between operator and machine. And we build that software inside the only environment where it can be properly proven — a live one, under real pressure, with a real enemy adapting against us every week.
The result is an R&D cycle that does not have release windows. It has engagements. Each one is a test the lab cannot run, because the lab does not have an opponent trying to break the code tonight.
Describing the loop from outside vs. living inside of it
This is the part that procurement struggles with, and it is structural rather than anyone’s fault.
You can fund a program. You can write a requirement, run a tender, and sign a contract. What you cannot purchase, fund into existence, or accelerate by any instrument is the operational knowledge that comes from thousands of real missions — the failure modes discovered the hard way, the jamming techniques met for the first time at the worst possible moment, the small adaptations that only make sense once you have watched a system fail in front of you.
Analysts across NATO now describe this in their own language — the short feedback loop, the OODA cycle at industrial scale, the fast-cycle space where Western procurement is structurally weakest. They are right, but…
There is a difference between describing the loop from the outside and living inside it. We are a Ukrainian-Latvian company. We do not study the transfer of frontline knowledge into allied frameworks as a topic. We are the transfer.
A system that has never operated in a contested EW environment carries unknown failure modes. A system that has been iterated through real engagements, with direct feedback from the people who depend on it, carries a different kind of reliability — the kind that no amount of laboratory validation can fully replicate.
The honest version
We are building toward something larger than any single platform: a software layer that lets one operator coordinate many autonomous systems at once, in conditions where most systems would go dark. That capability is maturing. We are not going to pretend it is finished.
But the loop that builds it is already running, and it is already the fastest part of what we do. That is the claim we are comfortable making at full volume — not what the system will be, but how it gets built. The proof of the future is the speed of the present.
The numbers point the same way. In 2025, a single unit operating our platforms neutralized targets worth roughly $6.1 million, at a leverage ratio of about 15 to 1, with a success rate near 74%. More than 90 frontline units field our systems. The team behind them includes 170-plus people, including those with direct combat experience. Those figures are not the loop — they are what the loop produces.
Why we are writing this down
Two reasons.
First, because the knowledge moves faster when it is shared deliberately than when it stays locked in a few heads. Late this May, we spent the better part of a week in Riga, across the defense-innovation events drawing the drone community to one place. Almost every serious conversation, on the exhibition floor and off it, came back to the same question: how do you keep a feedback loop running when the adversary is attacking the loop itself? Field Notes is part of our answer.
Second, because the future of autonomous defense will not be decided by who has the most impressive single platform. It will be decided by who can observe, adapt, and ship faster than the other side — and who can do it under conditions designed to make all three impossible.
We build systems that control assets at scale. We build the software that defines the advantage. And we build it where it counts.
More Field Notes to follow. Each one, a problem and the loop that closed it.
One operator. Dozens of autonomous systems. Software-defined.
Did this resonate?
If this resonates with how you’re thinking about defense capability — or if you’re working on similar problems and want to compare notes — we’d genuinely like to hear from you. We’re a team building something we think matters, and the conversations we learn most from are usually the ones we didn’t expect.
This article is part of the DK NEJET Field Notes series. Operational specifics are omitted or generalized to protect personnel, partners, and classified methods.