About the Role
Auric AI Labs is developing the world's most advanced AI-powered vision intelligence platform for defense applications. We are building an AI-native intelligence system from scratch. No legacy DSP pipeline, no AI-as-classifier-stage, no hybrid compromises. We need an engineer to build the deep learning models that detect, classify, geolocate, and fingerprint radar and communications emitters directly from raw IQ data.
Why this role exists
Most “AI in SIGINT” today is AI bolted onto a 1990s DSP pipeline as a final classification stage. Filter, threshold, extract feature, then hand the leftovers to a small neural network. Each stage before the network throws away information the network could have used. Each hand-crafted feature is a prior the model is forced to inherit.
We are not doing that.
We are betting that an AI-native architecture, where deep learning operates on raw IQ data with minimal pre-processing, learns its own features, and replaces the deterministic stack rather than supplementing it, delivers a 5-10x improvement on every metric that matters:
detection in low SNR, classification of unknown waveforms, specific emitter ID, real-time anomaly detection, cognitive EW.
This is not an “AI feature” inside a conventional product. The model is the product.
The actual problems
The signals are designed to defeat you: A contemporary LPI radar deliberately spreads its energy below the noise floor. A frequency-agile emitter changes frequency every pulse. Your job is to find structure that has been deliberately hidden.
You will not have real training data: Zero, for the signals that matter most. Before you can train a model, you have to answer a harder question: what does it mean for a model trained entirely in simulation to be trustworthy in the real world? You will design the synthetic data pipeline, the channel models, the validation methodology, and the test for knowing when your simulator is wrong.
Every emitter is unique: Two radars of the same model n