# EdgeLance 2026 Company Demo Narration

## 1. Mission intelligence for the team in the fight

EdgeLance is the mission operating layer for local artificial intelligence and disconnected teams. It brings models, mesh, sensors, devices, decisions, and evidence into one controlled system on the hardware operators already carry. The same signed mission truth supports each role before, during, and after the mission. Core workflows continue when the cloud is unavailable, synchronize when policy and connectivity allow, and preserve the evidence needed to understand what happened. This walkthrough uses synthetic demonstration data and current EdgeLance product screens. It does not imply Government validation or operational deployment.

## 2. Build the mission before the link disappears

Before launch, the team builds a Mission Pack. People, roles, map layers, standard procedures, intelligence files, approved models, retrieval libraries, and per-device loadouts are bound to the correct mission scope. Systems staff verify enrollment, software posture, mesh capacity, and rollback readiness. Analysts stage source material and confidence notes. Leaders resolve launch gates instead of discovering missing capability in the field. The same workflow supports training and rehearsal: change node count, sensor load, link quality, or the model loadout; run the scenario; review the result; then publish a signed package for disconnected delivery.

## 3. Fuse local cues into one operating picture

At a forward site or public-sector perimeter, EdgeLance fuses camera, radio-frequency, acoustic, identity, wearable, and motion cues into prioritized local tracks. Event-driven processing reduces compute and bandwidth while keeping the source, confidence, age, location, and synchronization state attached to every detection. Existing cameras and low-cost sensors can participate alongside mobile devices and specialist hardware. The operator sees what changed and why it matters, while the underlying clip or reading remains available for review. Local detection keeps the perimeter useful when reachback is slow, denied, or prohibited.

## 4. Recommendation, evidence, and authority stay together

EdgeLance turns mission context into decision support without turning a model into the decision maker. Every request passes through a policy-gated inference gateway that considers mission scope, classification, available capacity, and the approved model route. A course of action stays tied to the sensor evidence and sources that produced it. The operator can challenge the recommendation, inspect assumptions, select another action, or reject it. The resulting receipt records the model, quantization, policy decision, evidence, latency, and human disposition. Local models run on the node; base compute or an approved cloud is used only when policy and the link permit it.

## 5. The mission continues as the network changes

For a dismounted team, communications rarely stay on one transport. EdgeLance moves alerts, tasks, chat, clips, and mission state over the best link available, from satellite and local internet protocol networks down through radio, Bluetooth, low-rate mesh, or signed courier. Priority routing shapes content to the link, and store-and-forward catches up when teams reconnect. Live radio traffic becomes attributed, timestamped, searchable mission context while the recording and confidence remain inspectable. The signed mission ledger replicates peer to peer, so losing the command tablet or a relay changes the path, not the mission.

## 6. Shared context without ambiguous ownership

As the team moves, the operating picture follows across laptop, tablet, phone, watch, and approved partner surfaces. A drone feed, camera, radio net, or task can be handed from one operator to another with explicit control ownership. The transfer, identity, time, and acknowledgement enter the mission ledger, preventing double tasking and exposing an asset that has become unowned. Operators receive low-distraction actions on the device appropriate to the moment, while command sees the same mission state with a different view. EdgeLance coordinates the producing edge; it does not require an enterprise server to decide who owns the asset now.

## 7. Keep casualty, CASEVAC, and team context together

During a medical event, the medic opens a role-specific casualty view without leaving the shared mission. Vitals from approved wearables, blood type, allergies, injury context, location, medical supply readiness, and evacuation status remain together. The team lead sees operational impact without receiving unnecessary clinical detail. A CASEVAC request travels over the available mesh and keeps its acknowledgements when the network partitions. At transfer, EdgeLance exports a structured handoff record and preserves who observed, changed, and approved each item. The medic, operator, and commander work from one event while seeing only the information and actions appropriate to their roles.

## 8. Turn commercial hardware into managed mission nodes

Systems teams turn phones, tablets, laptops, cameras, wearables, relays, and base compute into managed mission nodes. Fleet Control shows operator assignment, verification, classification, mesh participation, software version, model loadout, and readiness. Signed packages move over the network or by USB-C courier, with staged rollout, health checks, and rollback when mission risk changes. Classification-aware policy and device management constrain what each node can receive. If a device is lost, duress and cryptographic zeroize can destroy the mission key and make protected payloads unrecoverable, while the content-minimized proof chain remains available for accountability.

## 9. Reconstruct the decisions that changed the mission

After action, EdgeLance reconstructs the timeline from signed events instead of asking the team to rebuild it from memory. Reviewers can replay sensor cues, radio traffic, chat, tasking, asset handoffs, artificial-intelligence recommendations, operator decisions, and linked evidence in mission time. Command can generate an after-action record; legal or partner reviewers can receive a scoped export; and analysts can turn validated observations into updated source notes, patterns, or Mission Pack content for the next operation. The feedback loop improves the mission layer while keeping raw evidence, interpretation, and human judgment distinct.

## 10. Local by default. Shared upward by choice.

EdgeLance connects upward without turning the enterprise into a runtime dependency. Approved tracks, tasks, incidents, evidence, changes, knowledge, and audit records can move into TAK-style systems, command platforms, intelligence workflows, or partner interfaces. A local GPU or Kubernetes environment can become preferred compute at a base while mobile nodes retain fallback capability. Zarf, UDS, and signed Mission Packs support air-gapped installation and disconnected delivery. The result is one operating loop: prepare, sense, decide, coordinate, act, protect, review, and improve. Different roles and devices; one mission truth. The operator remains the platform, and EdgeLance keeps the mission with the team.