Capture a request where the person is — then route it to the right department, with a trail.
That is the job. When a citizen or beneficiary needs something, you want to capture the request wherever they are — even on poor connectivity — and route it to the correct office, so you can respond with equity and traceability. Everythink draws your jurisdiction on the real map: the territory is the router.
The app the field worker holds — capture, route, place on the map.
This is the phone a field worker carries to a doorstep, a shelter or an intervention zone. Multimodal elicitation, routing to the owning department and jurisdictional rooms on the map are all in production today. We show the shape honestly — never a capability we have not built.
Illustrative — real images coming soon.
Requests arrive everywhere. They get lost between offices.
Here is how a program or relief operation actually runs today — and the topology that replaces it.
Requests come in through a counter, a phone line, and social media.
Citizen and beneficiary requests enter through the front desk, the phone, and social channels — and disappear in the gaps between departments. In the field, surveys are still done on paper and take weeks to systematize, long after the moment of need has passed.
Without jurisdictional routing, the loudest neighborhood beats the neediest.
When there is no map under the routing, the district with the most voice wins over the district with the most need. And without traceability there is no accountability that holds — not to the citizen, and not to the funder who is asking where the resources went.
Jurisdictional rooms on the real map — multimodal in the field, routed by territory.
Rooms drawn over the real map — district, neighborhood, intervention zone ✅ — with multimodal field elicitation by audio, image and geolocation across the engine's 13 question types ✅. Requests route to the correct department by geography and meaning, and every automated action is auditable and contestable by requirement, not by option (P-A6 / P-B5).
Scope commitment: civil and defensive applications only.
Everythink for government and humanitarian work serves civil and defensive applications only. No targeting. No offensive function. This is a written policy, published as part of the civil/defensive design (P-C9), not an aspiration — it is the boundary every deployment is held to. We would rather name the line plainly than leave it implied.
Your jurisdiction, drawn on the map — not buried in a queue.
Each level of the operation becomes a routable space with a real boundary, so requests land where the work happens.
Jurisdictional rooms Production
District, neighborhood, intervention zone — each is a room with its real polygon on the map. Routing is by geography and meaning, so a request reaches the office that actually owns it.
Explore the network →Multimodal field elicitation Production
Capture in the field by audio, image and geolocation across the engine's 13 question types — even on poor connectivity — so a survey is structured at the source instead of typed up weeks later.
Explore HAI Engine →
Request routing to the right department Production
Every captured request is placed by territory and intent and surfaced in that department's own panel — so nothing gets lost in the gap between the counter, the phone line and social media.
See routing & matchmaking →Capture, route and account for every request — with the honesty state on each line.
The core is in production today. The crisis-response and sovereignty pieces are named exactly for what they are — published design and roadmap, never upgraded.
- ✓Jurisdictional rooms on the real map — district, neighborhood and intervention zone as rooms with real polygons. ✅
- ✓Multimodal field elicitation — audio, image and geolocation across the 13 question types, built to work on poor connectivity. ✅
- ✓Request routing to the right department — placed by geography and meaning, reflected in the owning office's panel. ✅
- ✓Need↔resource matchmaking for crisis response — published civil/defensive design (P-C9); partial today, on the roadmap to full deployment. ⚠️ 🔵
- ✓Per-tenant data sovereignty & federation — public/private scope per room, federated across tenants (P-B6 / P-A7); a roadmap design. 🔵
- ✓Auditability and contestability as a requirement — every automated action can be reviewed and challenged (P-A6 / P-B5). ✅
Why a jurisdictional map beats a shared inbox.
Each capability keeps its honesty state. The status glyph sits with the feature — production, partial or roadmap, never rounded up.
| Feature | Advantage | Benefit |
|---|---|---|
| Jurisdictional rooms on the map ✅ | District, neighborhood and zone are rooms with real polygons. | A request is placed by territory, so the neediest district is visible, not drowned out. |
| Multimodal field elicitation ✅ | Audio, image and geolocation across 13 question types, built for poor connectivity. | Surveys are structured at the source, not transcribed from paper weeks later. |
| Request routing to the right department ✅ | Placement by geography and meaning into the owning office's panel. | Nothing falls through the gap between counter, phone line and social media. |
| Need↔resource matchmaking (crisis) ⚠️ 🔵 | Civil/defensive matchmaking design, published as P-C9. | A path to pair declared needs with available resources — partial today, roadmap to full. |
| Per-tenant sovereignty & federation 🔵 | Public/private scope per room, federated across tenants (P-B6 / P-A7). | Each agency keeps its own data and rules while still cooperating — a roadmap design. |
| Auditability & contestability ✅ | Every automated action is reviewable and challengeable (P-A6 / P-B5). | Accountability you can defend to a citizen and to a funder, by requirement. |
Crisis response and data sovereignty — designed and stated, not yet shipped.
We publish the design with an academic foundation before it becomes a commercial promise. Two of these pieces are not production today — the slots below are concepts, not shipped screenshots.
Need↔resource matchmaking for crisis Partial Roadmap
Pairing declared needs with available resources during crisis response, built on a published civil/defensive design (P-C9). The matchmaking core is partial; full crisis deployment is on the roadmap. No targeting and no offensive function — by written policy.
Explore matchmaking →Per-tenant sovereignty & federation Roadmap
Each tenant owns its data, with public or private scope set per room and federation across agencies (P-B6 / P-A7) — so departments cooperate without surrendering their own records or rules. A roadmap design, published before it is a shipped capability.
See federation context →These two are not built yet.
Need↔resource matchmaking for crisis response is partial — the matchmaking is modeled and the civil/defensive design is published as P-C9, but full crisis deployment is roadmap. Per-tenant sovereignty and federation (P-B6 / P-A7) is a roadmap design. We name the exact maturity rather than imply a capability we have not shipped. Want early access? Join the waitlist or read the underlying work in our academic papers.
The pieces a program or relief operation composes.
Start with the jurisdictional map and field capture, then add the modules each program needs.
Whitelabel Network Production
The branded shell over your whole jurisdiction — your name, your districts and zones, all on one map.
Explore Whitelabel →
HAI Engine Production
The engine behind multimodal field elicitation — 13 question types across audio, image and geolocation, built for poor connectivity.
Explore HAI →Matchmaking Partial
Need↔resource pairing for crisis response on the published civil/defensive design (P-C9) — partial today.
Explore Matchmaking →Respond with equity and traceability — from the field to the right office.
Capture a request wherever the citizen or beneficiary is, route it to the correct department by territory and meaning, and account for every action. Civil and defensive applications only — no targeting, no offensive function, in writing.