NASA Bearing Scout
2.6 KB model (Lean mode) — bearing fault classification.
What BitTrace Is
BitTrace delivers deterministic edge models you can actually ship. We build deployable “Scout” models—kilobyte-scale symbolic classifiers for constrained devices—designed for systems where power, footprint, and verification burden dominate. We operate like a model foundry (model factory): prove fit, build to constraints, then license for production.
- Symbolic, deterministic models
- Packed-bit computation
- Single-digit kilobyte models for real-world binary classification
- Edge deployment on MCUs / low-power CPUs / FPGAs / ASICs
- No floating point
Use Cases & Requirements
Deterministic packed-bit models remove floating-point drift and keep power budgets tight, enabling repeatable inference where resources and trust boundaries are constrained. Deterministic logic simplifies verification, reduces drift in industrial loops, and runs where floating point is unavailable.
- Aerospace / defense edge sensing
- Industrial monitoring
- IoT and microcontrollers
- Clinical monitoring and embedded health sensing
- Environments where deterministic behavior and low power are critical
Deterministic behavior
No floating-point pipeline
Low memory and power budget
Demonstrations
Bit-exact symbolic inference running on an ESP32-S3 microcontroller.
Heart Scout
Confidence-gated accept/reject behavior for uncertain inputs.
Built to ship
- Deterministic + testable (bit-exact parity)
- Few-kilobyte footprint for constrained devices
- Patent pending. Evaluation materials available under NDA.
Build your deterministic edge solution
Shipping on constrained hardware? We help teams deploy kilobyte-scale symbolic classifiers when power, footprint, and testability matter.
Contact us to discuss fit, timelines, and next steps:
info@bittrace.aiFull evaluation materials available under NDA.