Technology
AEVI-1
Libraries of candidate materials in the space environment at once, watched as they fail, learning what survives. In development.
What the platform does
64 materials per run
Whole composition libraries in parallel. The best existing platforms test one sample at a time.
Space-representative environment
High vacuum, thermal cycling −150 °C to +150 °C. The full LEO envelope, plus margin.
Thousands of cycles, unattended
Multi-week campaigns of a thousand cycles and more, closing the gap between published testing and orbit.
Crystal structure, live
In-situ XRD tracks each sample's phase during stress, not just before and after.
Full degradation picture
Optical imaging, per-device IV under illumination, and outgassing chemistry. The failure modes flight experiments found and ground tests miss.
AI in the loop
Every measurement feeds one ML-ready dataset. Models learn what survives and choose what to test next.
Why hasn't this been done?
It sits between fields. Space testing, crystallography, robotics, and ML each hold one piece.
The Real Product
Data that exists nowhere on Earth
No model can predict how a solar material degrades under coupled space stressors, because the training data has never existed.
Every AEVI-1 campaign records phase, optics, electrical output, and chemistry, cycle by cycle, across dozens of compositions. The dataset compounds with every run. It is where the value accumulates.
Have materials that need to survive space?
We are designing the first screening campaigns with partners. Send candidate compositions; get survival data nobody else can generate.
Talk to Us About Screening