An engineering company for spectrum and AI work
We build EW, GPS-free navigation, adversarial AI, and operator training for U.S. and allied units.
Who we are
Small company. Offices in California, Colorado, New York, and Estonia. We measure things before we ship them.
| Field | Detail |
|---|---|
| Locations | ca • co • ny • ee |
| Customers | U.S. and allied units working in jammed or GPS-denied environments |
| Spectrum | IQ compression, classification, transport, edge security, orchestration |
| AI | Knowledge collection, model realignment, authorized adversary emulation |
| Training | Bootcamps, simulation range, red-team engagements |
| Inquiries | Email inquiries@starzl.com |
Product Lines
Compresses IQ, classifies emitters, and sends the result over a constrained link.
SpectrumPosition and timing from ambient RF. No GPS, no transmissions. Same feed maps nearby emitters.
NavigationRed-team tools for models and networks. The same stack is used on infrastructure defense.
AIMulti-day courses on AI, cyber, and OSINT. The skills stay useful when the software changes.
TrainingHow we work
Start from the job, measure the system, keep a person in the loop on anything autonomous.
Write down the environment, the constraint, and the outcome before picking a product.
Agree on numbers for compression, latency, resilience, traceability, or source quality before anyone treats the system as ready.
Human review, source logs, and authority boundaries stay visible in AI workflows.
How the RF stack is split
EW and RF work is split into layers. Each layer has a job and a named component.
| Layer | Function | Component |
|---|---|---|
| 4 — Fusion & Effects | Sensor fleet orchestration and effects delivery | Odin (orchestration) & Mjolnir (effects) |
| 3 — Positioning & Navigation | Passive PNT and emitter localization | Starzl PNT |
| 2 — Signal Processing | Compression, classification, encryption | Starzl EW core |
| 1 — Network Transport | Jam-resistant RF link | Bifrost |
How we usually start
A technical conversation, a product evaluation, a prototype plan, or a discussion that needs a secure channel.
What gets better if the system works, which product addresses it, and what has to be measured before anyone should trust it.
Pick the product, then scope an evaluation against real data and a timeline.
Agree on the demo setup, the measures, and the review points. Move anything sensitive to the right channel.