The carrier of nothing.
The network of everyone.
Software-only wireless coverage built on top of every device that already exists. ~6.8B smartphones, 1.5B WiFi routers, 100M+ EVs, 4M+ satellites overhead — they relay each other's bytes; Loadit pays them per byte; HQ routes traffic; Hylaq is the identity layer underneath. No towers, no SIMs, no carriers.
live · iOS + Android · 30-second install
Install Hylaq Air on your phone
The .mobileconfig profile (iOS) + APK (Android) that puts every byte your phone sends through Hylaq Edge — DNS, tracker blocking, predictive prefetch, and the full guide for replacing "AT&T" with "Hylaq" on your device. The carrier of nothing, on YOUR phone.
year-3000 · proposed · pledge a node
Hylaq 6G — the carrier is the crowd
5G belongs to towers. 6G belongs to phones. Every Hylaq Air install is a soft-cell for the phones around it. $0 spectrum, $0 hardware, $0 partnerships — every iPhone already has the radios. Pledge your phone as a node.
new · year-4000 transport
Phantom mode — Hylaq is the carrier
No SIM, no WiFi, no Bluetooth, no LoRa. Phone screens flash bytes as light; another phone's camera reads them. Pure photons — you can see the data move.
The paradigm break
The hardware for a global wireless network is already deployed. ~6.8B phones each have 4 radios (cellular, WiFi, BLE, UWB). ~1.5B WiFi routers sit in homes worldwide. 100M+ Teslas have cellular + WiFi + UWB. AST SpaceMobile + Lynk now do direct-to-handset LEO. The hardware is everywhere. What's missing is the protocol that makes it cooperate as one network.
Carriers' moat was always rent on a coordination failure. Once relaying becomes an income stream for billions of edge devices, towers become legacy. The marginal cost of a Hylaq Air byte is somebody else's already-paid-for radio. Carriers physically cannot match that.
What we replace
Verizon + AT&T + T-Mobile
$220B/yr US cellular revenue
The 30-50% rake on a service that's mostly idle hardware. Air's per-byte pricing undercuts even prepaid plans, with no SIM and no contract.
Vodafone + China Mobile + Reliance
$1.7T/yr global cellular
Emerging-market carriers built towers their countries can't afford. Air uses the phones that are already there. The gateway-relay primitive turns each existing cellular subscriber into a cooperative ISP for their block.
Roaming agreements
$50B/yr extracted on travelers
Roaming is a coordination failure between carriers. With handle-native nodes, your phone is automatically discoverable in any country. Bytes flow at local Air rates.
Helium + DePIN coverage
$2B+ market, growing 100% YoY
Helium proved the demand for owner-operated coverage. Air does it without crypto-token speculation: you earn USD per byte, paid by Loadit, with no token to dump.
LEO direct-to-handset (AST, Lynk, Starlink Direct)
$30B+/yr by 2030
Air doesn't compete with LEO; it incorporates it. When peers can't reach, fallback to LEO bytes purchased wholesale + resold at protocol cost.
SIM cards + IMEI + roaming agreements
$5B+ legacy stack
Your @handle is your address. No SIM swap risk, no eSIM provisioning, no IMEI lock. Move between devices in 30 seconds.
Five devices. One network.
Phone on the charger
Idle 90% of the time. With Air running, your phone earns ~$30/mo relaying bytes for nearby phones via WiFi Direct + BLE. Your charger pays for itself.
Home WiFi router
80% headroom most days. Becomes a paid relay for your block. Your ISP doesn't care; you keep the difference. ~$50/mo passive income from the box already in your closet.
Tesla in a parking lot
WiFi + cellular + UWB sitting unused. Becomes a hyperlocal relay node for the whole lot. Owners earn while parked. Multiply by 100M+ Teslas globally.
Coffee-shop laptop
Most laptops have spare WiFi headroom + BT they never use. Become a gateway relay during your espresso. Earn back the price of the coffee.
LEO fallback
When no peers are reachable (rural, ocean, sky), Air buys LEO bytes wholesale (AST/Lynk/Starlink-Direct partnerships) and resells at protocol cost. Coverage everywhere on Earth, by default.
The next billion online
3B+ humans don't have affordable cellular. Air gives them peer-to-peer-to-LEO connectivity for a fraction of carrier prices. The next-billion-online story has been waiting for this rail.
How a packet moves
// Client requests a route:
POST /api/air/routes/request → Dijkstra over live links
// Path returned:
@alice/iphone ──[BLE]──▶ @neighbor/router
@neighbor/router ──[WIFI_AP]──▶ @stranger/laptop
@stranger/laptop ──[WIFI_DIRECT]──▶ @bob/macbook
// total: 0.04¢/MB locked, 55ms latency, 95% reliability
// Session settles:
settleSession() → Settlement Grid → relay payouts
Shipped today
- src/lib/air/ — types (8 RadioKind, 6 DeviceClass), capability scorer, reputation, per-radio spot pricing, in-memory node registry with heartbeat-TTL OFFLINE detection, link topology with TTL'd advertisements, Dijkstra pathfinder over the live link graph, session lifecycle (REQUESTED → ACTIVE → COMPLETED → SETTLED), settlement via Settlement Grid, worker SDK with httpTransport + airRelaySdk + runRelayLoop, BigInt-safe wire serialize
- 8 API routes — POST /nodes/register, POST /nodes/heartbeat, POST /links/advertise, POST /routes/request, POST /sessions/start, POST /sessions/complete (immediate Settlement Grid payout), GET /prices, GET /health
- Pricing curve — piecewise-linear utilization (floor → 4x → ceiling) per radio class. WIFI_DIRECT/AP at 0.01¢/MB floor, BLE 0.05, UWB 0.02, CELLULAR 0.50 (gateway-relay), LEO_SAT 1.00 (wholesale satellite). Air always undercuts even prepaid carrier rates.
- Capability scorer — per-class score caps (PHONE 350, LAPTOP 500, ROUTER 800, VEHICLE 600, SAT_GATEWAY 1000) weighted by radio kind (BLE 0.2x, WIFI_AP 1.2x, LEO_SAT 1.5x), RAM bonus, upstream-headroom bonus, sustained-power multiplier
- Reputation — bounded [0, 10000], +SUCCESS / -FAILURE / -SLASH, banned <1000, high-stakes-eligible >=3000. Pathfinder skips banned intermediate hops
- Phase-2 swap points called out — Postgres persistence, real radio adapters on iOS/Android/router firmware, LEO satellite gateway partnerships, fine-grained spectrum leases (idea #94), e2e + per-hop encryption
Carriers' 95% margin came from owning towers. We don't need towers.
~6.8B phones, 1.5B WiFi routers, 100M+ EVs, 4M+ LEO satellites already deployed. Hylaq Air is the protocol that turns them into one network — owned by the people running them.