Proving autonomous space systems
survive a contested fight.

We work on the simulation and stress-testing environments where resilient SATCOM and autonomous decision systems can be trained, red-teamed against jamming and electronic warfare, and validated before they touch a live network.

Get in touch → Areas of focus
Areas of focus

Where we can help.

01 · Simulation

Contested-environment simulation

Network-level digital twins of multi-orbit, hybrid (LEO / GEO / terrestrial) communications.

02 · Autonomy

Autonomous resource & spectrum coordination

Routing, link / beam / band allocation, and spectrum maneuver.

03 · Validation

Validation & red-teaming

Stress-testing decision systems against worst-case scenarios.

Research

Grounded in peer-reviewed work on
multi-agent learning for shared spectrum.

arXiv · 2503.24296 · March 2025
Fair Dynamic Spectrum Access via Fully Decentralized Multi-Agent Reinforcement Learning
Read the paper ↗ Download PDF
Founder

Trevor Gordon

ML research & engineering — ex-Google, ex-Apple. MS, Columbia University. Background spans applied multi-agent reinforcement learning, sensor systems, and production ML deployment.

LinkedIn ↗ arXiv ↗

If you're interested in partnering, please contact trevor@zamnaspace.com.