Beyond mesh
Traditional mesh systems were not designed for river corridors, moving equipment, or rapidly changing RF. FrogNet’s discovery and routing are built for clusters that split, move, and rejoin.
FrogNet · BullFrog LDC™ · AI Hosts
FrogNet is a portable, self-forming intranet and AI fabric for real environments – farms, river corridors, substations, shelters, and remote communities. It assumes WAN will fail, radios will be ugly, and power will be intermittent – and it keeps working anyway.
Why this exists
Most networking and AI assumes the same things: stable upstream connectivity, cooperative infrastructure, and acceptable latency. FrogNet assumes the opposite. It treats fracture, motion, and ugly RF as normal.
Traditional mesh systems were not designed for river corridors, moving equipment, or rapidly changing RF. FrogNet’s discovery and routing are built for clusters that split, move, and rejoin.
Most IoT platforms collapse when their vendor cloud disappears. FrogNet keeps services and logic at the edge, where your sensors, actuators, and operators actually live.
FrogNet AI Hosts are peers in the network, not just endpoints. They reason over local and shared state via the transient database, and collaborate across FrogNets when links allow.
What it is
A FrogNet is a cluster of small Linux nodes – Pis, NUCs, industrial PCs – connected by whatever radios and links you have. It discovers itself, forms an intranet, hosts applications and AI, and uses semantic compression to move only what matters.
Scenarios & pilots
FrogNet is a general fabric. Sensors and AI change by sector; the underlying capabilities remain the same. The book goes deep; pilots make it real in your world.
Suspended sensors and AI Hosts watching rising water levels can trigger local alerts even if every WAN link is gone. FrogNet treats the river corridor as its own sovereign network.
Large properties and irrigation districts become regions of cooperating FrogNets: soil sensors, pumps, pivots, and operators sharing a local intranet and AI even when broadband can’t reach.
Hazardous environments get a fabric that doesn’t quietly fail when the wrong switch flips or a link drops. Safety, monitoring, and response logic live inside the FrogNet itself.
I’m not looking for mass adoption yet. I’m looking for a small number of serious partners – in emergency management, agriculture, hydrology, utilities, and rural infrastructure – who see what this is and want to help shape how it rolls out into the world.
If you own a real-world problem that looks like this, we can design a pilot that starts from the running FrogNet codebase described in the book and adapts it to your terrain, radios, and constraints.
Talk about a pilot ➜See it in motion
Short walkthroughs showing FrogNet behavior, AI Hosts, and the dashboard. These can be swapped or expanded as new videos are published.
Overview of what a FrogNet is, why it exists, and how it behaves.
Demo of the AI Host – local reasoning with Knowledge Packs.
Dashboard view into the expert system and soft sensors.
Get involved
Whether you’re a potential pilot partner, investor, researcher, or operator, the first step is a conversation about your terrain, your constraints, and what “success” would mean.