Orbital AI Factory Node Architecture

An Orbital AI Factory isn't a single spacecraft – it's a cluster of high-power nodes working together in orbit.

Each node is a 1 MW-class AI data center in space: a central lattice spine with ROSA-style solar wings on each end, deployable radiators, and a chain of modular compute pods. A node carries everything it needs to run large AI models at the edge of space: power generation, thermal rejection, storage, networking, and high-bandwidth free-space optical (FSO) links.

Factories are created by linking multiple nodes into a coherent cluster:

• Nodes fly in coordinated orbits around the same target (a telescope, sensor constellation, station, or in-space factory).
• High-capacity FSO links between nodes form a single high-throughput fabric, so workloads, checkpoints, and datasets can move across the cluster as if it were one logical data center.
• Each node remains fully independent and serviceable: pods, wings, and radiators can be added, swapped, or upgraded without disturbing the rest of the factory.

This architecture scales like cloud regions: one node for early missions, then two, four, eight nodes as power and AI demand grow – all while reusing the same qualified hardware building blocks.

Modularity

Standardized pod rails with mechanical + power + data + coolant connectors

Grapple fixtures and clear approach corridors for tugs or station arms

Node designed to be assembled and upgraded in orbit, allowing for incremental capacity increases and technology refresh cycles without complete replacement.