UDS Fleet solves a real problem that everyone else ignores
Defense Unicorns launched UDS Fleet on June 2, 2026, purpose-built for deploying and managing software across distributed tactical systems. Fleet Connect ships as an Android app with a browser-based desktop client and iOS planned. DU demonstrated continuous delivery on an F-22, one of the most constrained computing environments in DoD.
Software delivery to tactical devices is genuinely hard. Air-gapped networks cannot pull from container registries. Classified environments require chain-of-custody for every binary. Disconnected devices need store-and-forward update mechanisms. DU built UDS to solve exactly this, backed by a $300M DoD-wide contract and a $136M Series B at unicorn valuation. EdgeLance does not try to solve the delivery problem because DU already has.
Delivery is infrastructure. Operations are what the operator cares about.
UDS Fleet gets a container, model weight file, or application binary onto a tactical device reliably and with provenance. That is the delivery layer. It does not define what that software does once running. It does not provide local AI inference, mission-aware routing, sensor fusion, or evidence capture. Those are operational capabilities above the delivery layer.
Getting a container onto an F-22 is impressive infrastructure engineering. Running local threat assessment on that F-22's sensor feeds without reachback is a different kind of software entirely.
Two layers, nobody owns both
Layer one: get software to the device. UDS Fleet, Zarf, and the broader DU ecosystem handle this. Containerized delivery, air-gap support, signing, provenance, fleet-wide update management.
Layer two: make the software fight. Local inference on the device hardware. Mesh networking when the cloud is unreachable. Sensor fusion across cameras, microphones, and RF receivers. Mission state that persists through reboots and link drops. Evidence capture with chain-of-custody. Ephemeral mission containers that wipe clean.
Neither layer works without the other. The best operational software is useless if you cannot get it to a classified device. The best delivery pipeline is useless if the software it delivers cannot operate independently at the edge.
What EdgeLance runs after Fleet delivers it
An EdgeLance mission pack arrives on a device as a signed bundle: model weights, mission configuration, mesh credentials, evidence schema. The capacity policy engine evaluates the device hardware profile and loads appropriate models. The mesh router establishes connections across available links. The evidence store initializes encrypted containers.
From there, the device is an autonomous tactical node. It runs inference on camera feeds, fuses sensor data, publishes entities via Cursor-on-Target for TAK interoperability, and captures every detection with provenance metadata. All without a connection to any cloud or operations center.
When the mission ends, the operator decides what to promote upward and what to destroy. The delivery layer does not need to know any of this. It delivered the software. EdgeLance made it fight. A MacBook command node with mission packs delivered via UDS Fleet through a Zarf air-gap bundle is a complete stack from repository to mission completion.