PaaSat is building satellite connectivity that lives inside the device, carried by a chip rather than bolted to a roof.
Satellite broadband asks you to install hardware and point it at the sky. That works for a house. It does not work for a motorbike, a water pump, a shipping crate, or a person walking between villages.
2.2 billion people still have no internet access. In Sub-Saharan Africa, only 37% of people are online.
Those are the ITU's latest figures, and they describe a problem the industry keeps answering with the wrong product: a dish, an installer, a monthly bill, and a fixed address.
Meanwhile the machines that would make rural economies work go dark the moment they leave a tower's range. Irrigation controllers. Delivery bikes. Cold chain boxes carrying vaccines. Water quality sensors. A gig worker's phone, three villages past the last mast.
None of those can carry a dish. All of them could carry a chip.
This is the part most explanations skip. Moving from a dish to a chip does not simply shrink the hardware. It changes what the link can carry, and therefore what the service is for. We would rather be straight about that than sell you a number we cannot hit.
A parabolic antenna concentrates a satellite signal roughly 3,000× more than a chip scale antenna can. That gain is what makes broadband possible.
A chip antenna has almost no gain, so the link must be narrow. We close it with orbit and spectrum instead of hardware, and we don't pretend it's broadband.
So we don't quote peak speeds. A chip link carries messages, telemetry, position and small payloads. It does so reliably, anywhere, on a battery that lasts. If you need to stream video to a farmhouse, buy a dish. If you need ten thousand devices that never go dark, that's us.
This is how the network operates. If you want the practical version, what setting up and using a device actually looks like, see how it works for you.
Not proprietary silicon. 3GPP Release 17 standardised satellite IoT (IoT-NTN), and modules already ship from MediaTek and Qualcomm. We are building the network those chips talk to, rather than asking anyone to adopt our hardware.
L-band and S-band. Lower frequencies punch through weather, foliage and buildings far better than Ku or Ka, and the radios are cheap and sparing with power. There is very little spectrum down there, which is exactly why the channel stays narrow.
Closer means less path loss, which is how you close a link to an antenna with no gain. It also means each satellite is overhead briefly, so coverage is a function of constellation design rather than raw power.
Traffic lands at a gateway, joins the terrestrial internet, and reaches your systems through an ordinary API. From your side it should look like any other device feed, just one that works where there is no tower.
Every one of these is a thing that currently stops working the moment it leaves coverage, and none of them can carry a dish.
Soil moisture, irrigation valves, tank levels, livestock trackers. Fields are, by definition, not where the towers are.
Motorbike couriers, long haul trailers, container seals. The route matters more than the depot, and the route has gaps.
Cold chain monitors for vaccines, remote clinic telemetry, emergency beacons. The failures here are measured in lives, not packets.
Solar installations away from the grid, borehole pumps, coastal generation. Assets deployed precisely where infrastructure isn't.
Gig workers who can't be verified, dispatched or paid because they're outside coverage. Connectivity as the precondition for income.
Research teams, fishing fleets, conservation patrols, reporters. People whose job is to be where the network isn't.
Every large operator optimises for the markets that pay best. Emerging markets are a rounding error to them. Here they are the entire thesis.
That means designing to a price a device operator can actually afford each month, not to a premium tier. It is not a marketing line. It is the constraint that decides whether a network gets used or merely admired.
The first deployments will be inside our own group, across gig work, energy and field reporting. Being our own first customer is how we find out what breaks before anyone else has to.
Satellite companies are unusually prone to describing plans as products. This page is the antidote. It is dated, it is specific, and it includes the parts that are not finished.
Last updated 10 August 2026. If something on this list stops being true, it changes here first.
If you have designed against 3GPP Release 17 IoT NTN, you have already done the hard part.
Certification is portable: a device certified for R17 IoT-NTN works on any compliant network. That means no lock in to us, and no bespoke integration work to try us. We think that is the only honest way to ask a hardware team to take a bet on a new operator.
Module families we are targeting for compatibility:
Device APIs and a sandbox will follow the pilot. If you are building hardware that needs to work outside coverage, tell us what it is. That shapes what we build first.
There is nothing to buy yet, and we would rather not pretend otherwise. But if you operate devices that go dark, we want to hear what they are. That decides what we build first.